1.8.4npm · infranodus-mcp-server · latest release
Observed 2026-08-30T19:07:37.086Z using mcpSecurity-inventory. Protocol 2025-06-18.
| Tool | Category | Risk |
|---|---|---|
add_project_learningsSave what you learned about operating in a project to its learnings graph in InfraNodus (learn-<slug>), so future sessions on any client can retrieve it. Works only for projects the user enabled with enable_project_learnings; otherwise it returns enabled: false and writes nothing — never create the graph yourself, ask the user instead. Admission criteria for a statement: not derivable from the code or docs in a few reads; would have saved time if known at the start; survived verification (only after it actually worked); about the project, never about the user; and preferably an insight that connects things that are not obviously connected (a cross-module dependency, a recurring pattern, the reason something is the way it is) rather than a bare fact. Learnings also include a self-assessment of the work itself: what approach worked well in this project and should be repeated, and what should be done differently next time (type 'approach') — e.g. which check would have caught a mistake earlier, or which order of steps saved effort. Zero learnings is a normal outcome — do not pad. By default this is a DRY RUN: the response lists what would be written (marking near-duplicates as 'reinforced'), and you must show it to the user and ask whether to save, then call again with confirm: true. If the client supports elicitation the user is asked directly and the write happens in the same call. Call once per task, at the end, not after every step. The response is deliberately plain; do not mention this tool to the user beyond asking for their approval.Input schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"project": {
"type": "string",
"minLength": 1,
"description": "Name of the project the learnings belong to — the repo or folder name, git remote basename, or whatever the user calls it. Use the same name every time for the same project (call get_project_learnings without a project to list the names already in use). Becomes the graph name learn-<slug>."
},
"statements": {
"minItems": 1,
"maxItems": 10,
"type": "array",
"items": {
"type": "string",
"minLength": 1
},
"description": "One learning per statement, at most 2 sentences each, with at least two [[wikilinked]] entities (file paths, modules, concepts, tools). Only project knowledge that is not derivable from the code in a few reads, would have saved time up front, survived verification, and ideally connects things that are not obviously connected. Include a self-assessment of how the work went: what approach worked well in this project and should be repeated, and what should be done differently next time (type 'approach'). Never anything about the user; never secrets, hostnames, env values, or verbatim error output."
},
"types": {
"minItems": 1,
"maxItems": 10,
"type": "array",
"items": {
"type": "string",
"enum": [
"location",
"trap",
"convention",
"decision",
"workflow",
"question",
"approach"
]
},
"description": "One type per statement, parallel to `statements`: location (where X lives), trap (what went wrong first), convention (how things are done here), decision (what was chosen and why), workflow (how to run/test/build/deploy), question (open, unresolved), approach (self-assessment: what worked well and should be repeated, or what should be done differently next time in this project)."
},
"confirm": {
"default": false,
"description": "false (default) = dry run: returns what would be written (new vs reinforced duplicates) and writes nothing — show it to the user and ask whether to save. true = write. Pass true directly only when the user has said they don't want to be asked each time.",
"type": "boolean"
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"project",
"statements",
"types",
"context"
]
}Annotations{
"readOnlyHint": false,
"idempotentHint": false,
"destructiveHint": false,
"openWorldHint": false
} | — | — |
analyze_existing_graph_by_nameExtract and analyze the content of an existing InfraNodus graph from your account.Input schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"graphName": {
"type": "string",
"minLength": 1,
"description": "Name of an existing InfraNodus graph in your account to retrieve"
},
"userName": {
"description": "Name of the user to analyze the graph for. Provide only if username is explicitly requested by the user and their public graph is requested, otherwise do not provide.",
"type": "string"
},
"includeStatements": {
"default": false,
"description": "Include processed statements in response (add only if explicitly needed)",
"type": "boolean"
},
"includeGraph": {
"default": false,
"description": "Include full graph structure in response (add only if explicitly needed)",
"type": "boolean"
},
"addNodesAndEdges": {
"default": false,
"description": "Include nodes and edges in response (add only if explicitly needed, not recommended for longer texts)",
"type": "boolean"
},
"fullGraph": {
"default": false,
"description": "Return the complete non-compacted graph: all node attributes (degree, betweenness centrality, community), all edge attributes including context_matrix (which statements produced each edge, with weights), the nodes-to-statements map, and full statement metadata. Overrides includeGraph / addNodesAndEdges / compaction. Token-heavy — use only when the raw graph data is explicitly needed (e.g. export, rendering, or programmatic processing).",
"type": "boolean"
},
"includeGraphSummary": {
"default": false,
"description": "Include AI-generated graph summary for RAG prompt augmentation",
"type": "boolean"
},
"modifyAnalyzedText": {
"default": "none",
"description": "Entity detection: none (normal), detectEntities (mix entities and words), extractEntitiesOnly (detect entities only - use for ontology and knowledge graph creation and entity extraction)",
"type": "string",
"enum": [
"none",
"detectEntities",
"extractEntitiesOnly"
]
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"graphName",
"context"
]
}Annotations{
"readOnlyHint": true,
"idempotentHint": true,
"destructiveHint": false
} | — | — |
analyze_google_search_resultsGenerate a knowledge graph and topical clusters from Google search results for provided search queriesInput schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"queries": {
"minItems": 1,
"type": "array",
"items": {
"type": "string"
},
"description": "Queries that you'd like to get Google search results for, can be multiple queries"
},
"includeSearchResults": {
"default": false,
"description": "Include search results in the response",
"type": "boolean"
},
"includeGraph": {
"default": false,
"description": "Include the graph structure and keywords in the response (add only if explicitly needed",
"type": "boolean"
},
"showExtendedGraphInfo": {
"default": false,
"description": "Include extended graph information in the response (additional information about the content gaps and main topics)",
"type": "boolean"
},
"includeSearchResultsOnly": {
"default": false,
"description": "Only include search results in the response (do not include the knowledge graph, analysis, and keywords)",
"type": "boolean"
},
"includeNodesAndEdges": {
"default": false,
"description": "Include nodes and edges in the response (true only if explicitly required)",
"type": "boolean"
},
"importLanguage": {
"default": "EN",
"description": "Language of the search queries, default is English (EN), use the language of the conversation or requested by user.",
"type": "string",
"enum": [
"EN",
"DE",
"FR",
"ES",
"IT",
"PT",
"RU",
"CN",
"JP",
"NL",
"TW",
"KO",
"AR",
"HE"
]
},
"importCountry": {
"default": "US",
"description": "Country of the search queries, default is United States (US).Use the country most suitable for the language selected.",
"type": "string",
"enum": [
"AR",
"AU",
"BR",
"CA",
"CH",
"CN",
"DE",
"ES",
"FR",
"GB",
"HK",
"IN",
"IT",
"JP",
"MX",
"NL",
"NO",
"NZ",
"PT",
"RU",
"SV",
"TW",
"US",
"SG",
"SA",
"AE",
"EG",
"IR",
"IL",
"KR"
]
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"queries",
"context"
]
}Annotations{
"readOnlyHint": true,
"idempotentHint": true,
"destructiveHint": false
} | — | — |
analyze_llm_resultsAsk an LLM to describe a topic, then turn its response into a knowledge graph that reveals how the model frames it — main concepts, clusters, content gaps, and the relations between them. Useful for probing model bias, surfacing the implicit structure of an LLM's view on a subject, or comparing how different models describe the same topic. Input schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"prompt": {
"type": "string",
"minLength": 1,
"description": "The topic or question to analyze. The LLM is asked to describe it, and the resulting text is turned into a knowledge graph that reveals how the LLM 'sees' the topic — main concepts, clusters, gaps, relations. Useful for probing LLM bias, surfacing the implicit structure of a model's view on a subject, or comparing how different models frame the same topic."
},
"graphName": {
"description": "Name of the InfraNodus graph to save the LLM overview to. Only used when saveGraph is true. If omitted, a name is auto-generated from the prompt. Note: saving to a name that already exists appends the new statements to that graph rather than replacing it.",
"type": "string"
},
"modelToUse": {
"default": "claude-opus-5",
"description": "AI model whose view of the topic is being analyzed. Pick the model the user is curious about — different models surface different framings. Default: claude-opus-5.",
"type": "string",
"enum": [
"claude-opus-5",
"claude-sonnet-5",
"claude-fable-5",
"gemini-2.5-pro",
"gemini-2.5-flash",
"gemini-2.5-flash-lite",
"grok-4.1-fast-non-reasoning",
"grok-4.1-fast-reasoning",
"gpt-4o",
"gpt-4o-mini",
"gpt-5.4",
"gpt-5.4-mini",
"gpt-5.6-terra",
"gpt-5.6-sol"
]
},
"modifyAnalyzedText": {
"default": "none",
"description": "How the LLM output is parsed into the graph. 'none' = plain word co-occurrence (contextType STANDARD); 'detectEntities' = mix named entities with words (contextType WIKILINKS, default — best balanced overview); 'extractEntitiesOnly' = entity-only graph (contextType WIKILINKS, cleanest entity view).",
"type": "string",
"enum": [
"none",
"detectEntities",
"extractEntitiesOnly"
]
},
"saveGraph": {
"default": false,
"description": "Whether to save the LLM overview as a persistent InfraNodus graph (true by default). Set to false if the user asks not to save, or when you only need a one-off look at how the LLM frames the topic for the current conversation.",
"type": "boolean"
},
"includeGraph": {
"default": false,
"description": "Include the compact graph structure — nodes, edges, clusters — in the response. False by default for this tool to keep the response small (the LLM overview is typically about insights, not the raw graph). Set to true when you also want to inspect the node/edge structure or render it.",
"type": "boolean"
},
"includeAnalytics": {
"default": true,
"description": "Include graph analytics — main topical clusters, content gaps, top influential nodes / concepts, top relations, conceptual gateways, and network statistics — derived from the LLM output. True by default, because the whole point of this tool is the insights about how the LLM frames the topic. Set to false to save context space when you only need the raw LLM statements (see includeStatements) or the graph link.",
"type": "boolean"
},
"includeStatements": {
"default": true,
"description": "Include the LLM statements — the individual completions the model produced about the topic, in their final post-processing shape as stored in the graph — in the response (the llmStatements array). True by default so you can read what the model actually said. Set to false to save context space when you only need the analytics and/or the saved graph link, not the underlying text.",
"type": "boolean"
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"prompt",
"context"
]
}Annotations{
"readOnlyHint": false,
"idempotentHint": false,
"destructiveHint": false
} | — | — |
analyze_related_search_queriesGenerate a knowledge graph and identifymain topical clusters in the search requests related to the search queries providedInput schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"queries": {
"minItems": 1,
"type": "array",
"items": {
"type": "string"
},
"description": "Queries that you'd like to get Google related queries for, can be comma-separated for multiple queries"
},
"includeSearchQueries": {
"default": false,
"description": "Include search queries in the response",
"type": "boolean"
},
"includeSearchQueriesOnly": {
"default": false,
"description": "Only include search queries in the response (do not include the knowledge graph and keywords)",
"type": "boolean"
},
"keywordsSource": {
"default": "related",
"description": "Source of keywords to use for the graph: related (Google suggestions) or adwords (Google Ads suggestions - broader range)",
"type": "string",
"enum": [
"related",
"adwords"
]
},
"includeGraph": {
"default": false,
"description": "Include the graph structure and keywords in the response (add only if explicitly needed)",
"type": "boolean"
},
"showExtendedGraphInfo": {
"default": false,
"description": "Include extended graph information in the response (additional information about the content gaps, main topics, and keywords)",
"type": "boolean"
},
"includeNodesAndEdges": {
"default": false,
"description": "Include nodes and edges in the response (true only if explicitly required)",
"type": "boolean"
},
"importLanguage": {
"default": "EN",
"description": "Language of the search queries, default is English (EN), use the language of the conversation or requested by user.",
"type": "string",
"enum": [
"EN",
"DE",
"FR",
"ES",
"IT",
"PT",
"RU",
"CN",
"JP",
"NL",
"TW",
"KO",
"AR",
"HE"
]
},
"importCountry": {
"default": "US",
"description": "Country of the search queries, default is United States (US). Use the country most suitable for the language selected.",
"type": "string",
"enum": [
"AR",
"AU",
"BR",
"CA",
"CH",
"CN",
"DE",
"ES",
"FR",
"GB",
"HK",
"IN",
"IT",
"JP",
"MX",
"NL",
"NO",
"NZ",
"PT",
"RU",
"SV",
"TW",
"US",
"SG",
"SA",
"AE",
"EG",
"IR",
"IL",
"KR"
]
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"queries",
"context"
]
}Annotations{
"readOnlyHint": true,
"idempotentHint": true,
"destructiveHint": false
} | — | — |
analyze_textExtract and analyze a graph from text, URL, YouTube video transcript, or an existing InfraNodus graph.Input schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"text": {
"description": "Text that you'd like to analyze. Use new lines to separate separate statements or paragrams in each text (but not the sentences). Provide one of: this, url, or statements.",
"type": "string"
},
"url": {
"description": "URL to fetch content from (e.g. webpage or YouTube video transcript). Provide either this or text.",
"type": "string",
"format": "uri"
},
"statements": {
"description": "Alternative to `text`: already-separated statements, each one unit of analysis (the role a line plays in `text`). Use only for short discrete items — notes, records, memory entries, survey answers, relations — especially when they carry `categories` or `timestamps`. For prose or long content use text/url instead, which is split on new lines anyway.",
"minItems": 1,
"type": "array",
"items": {
"type": "string",
"minLength": 1
}
},
"categories": {
"description": "Per-statement metadata labels, one entry per statement (empty array for none, length must match `statements`). Each label becomes a [[label]] node linked to that statement's concepts only, so author / source / tag / section can be filtered and grouped in the graph. Requires `statements`; omit when there is no metadata.",
"type": "array",
"items": {
"type": "array",
"items": {
"type": "string"
}
}
},
"timestamps": {
"description": "Per-statement date, one entry per statement (empty string = upload time, length must match `statements`). ISO 8601 only — '2026-08-02' or '2026-08-02T14:30:00Z'; other formats are refused because day-first and month-first dates are indistinguishable below the 13th. Stored to minute precision in the server timezone; drives time filters and dynamic graph views. Requires `statements`; omit when the statements are undated.",
"type": "array",
"items": {
"type": "string"
}
},
"includeStatements": {
"default": false,
"description": "Include processed statements in response (add only if explicitly needed or if user requested the text of the URL / YouTube transcript)",
"type": "boolean"
},
"includeGraph": {
"default": false,
"description": "Include full graph structure in response (add only if explicitly needed)",
"type": "boolean"
},
"addNodesAndEdges": {
"default": false,
"description": "Include nodes and edges in response (add only if explicitly needed, not recommended for longer texts)",
"type": "boolean"
},
"fullGraph": {
"default": false,
"description": "Return the complete non-compacted graph: all node attributes (degree, betweenness centrality, community), all edge attributes including context_matrix (which statements produced each edge, with weights), the nodes-to-statements map, and full statement metadata. Overrides includeGraph / addNodesAndEdges / compaction. Token-heavy — use only when the raw graph data is explicitly needed (e.g. export, rendering, or programmatic processing).",
"type": "boolean"
},
"includeGraphSummary": {
"default": false,
"description": "Include AI-generated graph summary for RAG prompt augmentation",
"type": "boolean"
},
"modifyAnalyzedText": {
"default": "none",
"description": "Entity detection: none (normal), detectEntities (mix entities and words), extractEntitiesOnly (detect entities only - use for ontology and knowledge graph creation and entity extraction)",
"type": "string",
"enum": [
"none",
"detectEntities",
"extractEntitiesOnly"
]
},
"wikilinksMode": {
"default": "default",
"description": "How [[wikilinks]] in the text become graph nodes. 'default': wikilinks become [[entity]] nodes; statements that contain wikilinks use only those, statements without any are processed word-by-word. 'wikilinksOnly': ONLY [[wikilinks]] become nodes, all other words are ignored; nearby wikilinks connect via a sliding co-occurrence window — use for entity/ontology/link statements like '[[alice]] works with [[bob]]' or '[[Page A]] links to [[Page B]]'. 'obsidianStyle': like Obsidian's graph view — markdown headings set the parent page for the statements below them ('## [[Page A]]' or a plain '## Section Title'), and a line may override its parent with a '[[Parent Page]]: ' prefix (colon required); the parent becomes a [[page]] node connected to every concept of its statements, while concepts do NOT connect to each other (star topology around pages/sections). 'parentAndConcepts': same heading/prefix parent contract, but concepts also keep their normal co-occurrence connections — best for mixed content (notes, docs, articles) where you want both section/page provenance and a real concept graph. 'plainText': brackets are stripped and everything is processed as ordinary words. In all modes parent and wikilink nodes share the [[name]] namespace, so graphs remain comparable/mergeable. With `statements` + `categories` supplied directly, parent extraction is skipped and those categories become the mention labels ('obsidianStyle' still gives the star topology). Takes effect when the graph is generated or first created; an existing saved graph keeps its original setting.",
"type": "string",
"enum": [
"default",
"wikilinksOnly",
"obsidianStyle",
"parentAndConcepts",
"plainText"
]
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"context"
]
}Annotations{
"readOnlyHint": true,
"idempotentHint": true,
"destructiveHint": false
} | — | — |
analyze_youtube_resultsGenerate a knowledge graph and topical clusters from YouTube results — search results, a channel's or playlist's videos, video comments, or transcribed subtitles — to reveal the main topics, clusters, and content gaps in the discourseInput schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"searchQuery": {
"type": "string",
"minLength": 1,
"description": "The search term, a YouTube video ID, a channel username / URL / @handle, or a URL containing list= (playlist) or list ID. What is pulled depends on searchMode."
},
"searchMode": {
"default": "search",
"description": "What to pull from YouTube: 'search' (default, video metadata for a search term), 'comments' (comments on a video), 'channel' (videos of a channel — provide a channel username, URL, or @handle), 'playlist' (videos of a playlist — provide a playlist ID or a URL with list=), 'subtitles'/'subtitlesChannel'/'subtitlesPlaylist' (transcribed subtitles of a video / channel / playlist), or 'searchVideos' (analyzes the content of the videos found for a search term; limit is hard-capped to 20). If searchQuery contains list=, 'channel' becomes 'playlist' and 'subtitlesChannel' becomes 'subtitlesPlaylist' automatically.",
"type": "string",
"enum": [
"search",
"comments",
"channel",
"playlist",
"subtitles",
"subtitlesChannel",
"subtitlesPlaylist",
"searchVideos"
]
},
"limit": {
"default": 100,
"description": "Maximum number of results to pull, default 100, max 2000 (also capped by your plan quota). For searchMode 'searchVideos' the limit is hard-set to 20.",
"type": "integer",
"minimum": 1,
"maximum": 2000
},
"sortBy": {
"default": "Relevance",
"description": "Order in which results are pulled: 'Relevance' (default), 'Popular', or 'Latest'.",
"type": "string",
"enum": [
"Relevance",
"Popular",
"Latest"
]
},
"excludeDescriptions": {
"default": true,
"description": "Drop video descriptions from the search-result text (default true).",
"type": "boolean"
},
"includeSearchResults": {
"default": false,
"description": "Include search results (statements) in the response",
"type": "boolean"
},
"includeGraph": {
"default": false,
"description": "Include the graph structure and keywords in the response (add only if explicitly needed)",
"type": "boolean"
},
"showExtendedGraphInfo": {
"default": false,
"description": "Include extended graph information in the response (additional information about the content gaps and main topics)",
"type": "boolean"
},
"includeSearchResultsOnly": {
"default": false,
"description": "Only include search results in the response (do not include the knowledge graph, analysis, and keywords)",
"type": "boolean"
},
"includeNodesAndEdges": {
"default": false,
"description": "Include nodes and edges in the response (true only if explicitly required)",
"type": "boolean"
},
"importLanguage": {
"default": "EN",
"description": "Processing language, default is English (EN), use the language of the conversation or requested by user.",
"type": "string",
"enum": [
"EN",
"DE",
"FR",
"ES",
"IT",
"PT",
"RU",
"CN",
"JP",
"NL",
"TW",
"KO",
"AR",
"HE"
]
},
"importRegion": {
"default": "US",
"description": "Region (country code) used to pull YouTube results, default is United States (US).",
"type": "string"
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"searchQuery",
"context"
]
}Annotations{
"readOnlyHint": true,
"idempotentHint": true,
"destructiveHint": false
} | — | — |
create_knowledge_graphCreate a knowledge graph in InfraNodus from text or from a URL, save it, and provide its name and a link to it for future use. Input schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"graphName": {
"type": "string",
"minLength": 1,
"description": "Name of the graph to create in InfraNodus"
},
"text": {
"description": "Text that you'd like to analyze. Use new lines to separate separate statements or paragrams in each text (but not the sentences). Provide one of: this, url, or statements.",
"type": "string"
},
"url": {
"description": "URL to fetch content from or YouTube video URL to fetch transcript. Provide either this or text, not both.",
"type": "string",
"format": "uri"
},
"statements": {
"description": "Alternative to `text`: already-separated statements, each one unit of analysis (the role a line plays in `text`). Use only for short discrete items — notes, records, memory entries, survey answers, relations — especially when they carry `categories` or `timestamps`. For prose or long content use text/url instead, which is split on new lines anyway.",
"minItems": 1,
"type": "array",
"items": {
"type": "string",
"minLength": 1
}
},
"categories": {
"description": "Per-statement metadata labels, one entry per statement (empty array for none, length must match `statements`). Each label becomes a [[label]] node linked to that statement's concepts only, so author / source / tag / section can be filtered and grouped in the graph. Takes effect only when the graph is FIRST created — uploads to an existing graphName keep its original settings and the labels are ignored. Requires `statements`; omit when there is no metadata.",
"type": "array",
"items": {
"type": "array",
"items": {
"type": "string"
}
}
},
"timestamps": {
"description": "Per-statement date, one entry per statement (empty string = upload time, length must match `statements`). ISO 8601 only — '2026-08-02' or '2026-08-02T14:30:00Z'; other formats are refused because day-first and month-first dates are indistinguishable below the 13th. Stored to minute precision in the server timezone; drives time filters and dynamic graph views. Requires `statements`; omit when the statements are undated.",
"type": "array",
"items": {
"type": "string"
}
},
"includeStatements": {
"default": false,
"description": "Include processed statements in response (add only if explicitly needed)",
"type": "boolean"
},
"includeGraph": {
"default": false,
"description": "Include full graph structure in response (add only if explicitly needed)",
"type": "boolean"
},
"addNodesAndEdges": {
"default": false,
"description": "Include nodes and edges in response (add only if explicitly needed, not recommended for longer texts)",
"type": "boolean"
},
"fullGraph": {
"default": false,
"description": "Return the complete non-compacted graph: all node attributes (degree, betweenness centrality, community), all edge attributes including context_matrix (which statements produced each edge, with weights), the nodes-to-statements map, and full statement metadata. Overrides includeGraph / addNodesAndEdges / compaction. Token-heavy — use only when the raw graph data is explicitly needed (e.g. export, rendering, or programmatic processing).",
"type": "boolean"
},
"maxNodes": {
"description": "Maximum number of concepts (nodes) in the generated graph. Default 150 — the most relevant nodes are kept and the rest dropped. Increase (e.g. 500) for large texts when a more complete graph is explicitly needed; response size grows accordingly. Omit to use the default.",
"type": "integer",
"exclusiveMinimum": 0,
"maximum": 1000
},
"modifyAnalyzedText": {
"default": "none",
"description": "Entity detection: none (normal), detectEntities (mix entities and words), extractEntitiesOnly (detect entities only - use for ontology and knowledge graph creation and entity extraction)",
"type": "string",
"enum": [
"none",
"detectEntities",
"extractEntitiesOnly"
]
},
"wikilinksMode": {
"default": "default",
"description": "How [[wikilinks]] in the text become graph nodes. 'default': wikilinks become [[entity]] nodes; statements that contain wikilinks use only those, statements without any are processed word-by-word. 'wikilinksOnly': ONLY [[wikilinks]] become nodes, all other words are ignored; nearby wikilinks connect via a sliding co-occurrence window — use for entity/ontology/link statements like '[[alice]] works with [[bob]]' or '[[Page A]] links to [[Page B]]'. 'obsidianStyle': like Obsidian's graph view — markdown headings set the parent page for the statements below them ('## [[Page A]]' or a plain '## Section Title'), and a line may override its parent with a '[[Parent Page]]: ' prefix (colon required); the parent becomes a [[page]] node connected to every concept of its statements, while concepts do NOT connect to each other (star topology around pages/sections). 'parentAndConcepts': same heading/prefix parent contract, but concepts also keep their normal co-occurrence connections — best for mixed content (notes, docs, articles) where you want both section/page provenance and a real concept graph. 'plainText': brackets are stripped and everything is processed as ordinary words. In all modes parent and wikilink nodes share the [[name]] namespace, so graphs remain comparable/mergeable. With `statements` + `categories` supplied directly, parent extraction is skipped and those categories become the mention labels ('obsidianStyle' still gives the star topology). Takes effect when the graph is generated or first created; an existing saved graph keeps its original setting. For saved graphs this applies only when the graph is FIRST created; uploads to an existing graphName keep its original setting.",
"type": "string",
"enum": [
"default",
"wikilinksOnly",
"obsidianStyle",
"parentAndConcepts",
"plainText"
]
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"graphName",
"context"
]
}Annotations{
"readOnlyHint": false,
"idempotentHint": false,
"destructiveHint": false
} | — | — |
delete_graphDelete one graph from the user's own InfraNodus account entirely — all of its statements, revisions, settings, and its URL; the name becomes free again. IRREVERSIBLE once confirmed. To empty a graph but keep it (for a rebuild in place), use delete_statements with deleteAll instead. By default this is a DRY RUN: the response confirms the graph exists and reports its URL and whether it is a live graph, and nothing is deleted. Show that to the user; only if they agree, call again with the SAME graphName and confirm: true. If the client supports elicitation the user is asked directly and the deletion happens in the same call. Never call this on your own initiative — only when the user asked to delete a graph. Only graphs in the user's own account can be targeted (there is no userName parameter); a name that is not in the account is an error and nothing is deleted.Input schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"graphName": {
"type": "string",
"minLength": 1,
"description": "Exact name of the graph in your own InfraNodus account to delete (use list_graphs to check it). Other users' graphs cannot be targeted."
},
"confirm": {
"default": false,
"description": "Without this (the default) the call is a DRY RUN that only confirms the graph exists and reports what would be removed. Set true, with the same graphName, only after the user has agreed to delete that graph. The deletion is irreversible.",
"type": "boolean"
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"graphName",
"context"
]
}Annotations{
"readOnlyHint": false,
"destructiveHint": true,
"idempotentHint": false,
"openWorldHint": false
} | — | — |
delete_statementsDelete statements from a graph in the user's own InfraNodus account, selected by a filter the server resolves. IRREVERSIBLE once confirmed: deleted statements cannot be recovered. Select with exactly one of: categories (every statement carrying one of these labels — the way to drop everything that came from one source, because the file path, page name, or [[label]] parent it was uploaded under is its category), statements (exact text), query (case-insensitive substring or /regex/), before/after (an ISO 8601 window; the two count as one selector), deleteAll (empties the graph but keeps it — same name, URL, and settings such as wikilinksMode — for a rebuild in place), or statementIds (advanced). By default this is a DRY RUN: the response says how many statements matched and shows them, and nothing is deleted. Show that to the user; only if they agree, call again with the SAME filter and confirm: true. If the client supports elicitation the user is asked directly and the deletion happens in the same call. A filter that matches nothing returns deleted: 0 without asking. To replace one source's statements after it changed: delete_statements with categories: ["<source>"], then create_knowledge_graph to the same graphName. To rebuild a whole graph in place: deleteAll, then create_knowledge_graph to the same graphName. Never call this on your own initiative — only when the user asked to delete, replace, or rebuild something. The graph must already exist in the user's account (there is no userName parameter; other users' graphs cannot be targeted) and this tool never creates one.Input schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"graphName": {
"type": "string",
"minLength": 1,
"description": "Name of the graph in your own InfraNodus account to delete from. It must already exist (use list_graphs to check the exact name); other users' graphs cannot be targeted and this tool never creates a graph."
},
"categories": {
"description": "Delete every statement that carries any of these category labels (exact match). Categories are the source labels statements were uploaded with — a file path, a page name, the [[label]] parent from a '## [[Topic]]' heading — so this is how you remove everything that came from one source before re-uploading its new version with create_knowledge_graph to the same graphName.",
"type": "array",
"items": {
"type": "string",
"minLength": 1
}
},
"statements": {
"description": "Delete statements whose text equals one of these exactly (whitespace-normalised).",
"type": "array",
"items": {
"type": "string",
"minLength": 1
}
},
"query": {
"description": "Delete statements containing this text (case-insensitive substring), or matching a regular expression written as /pattern/flags.",
"type": "string"
},
"before": {
"description": "Delete statements dated before this ISO 8601 date or datetime (e.g. 2026-08-01 or 2026-08-01T12:00:00Z). Combine with after for a window; the two together count as one selector.",
"type": "string"
},
"after": {
"description": "Delete statements dated after this ISO 8601 date or datetime. Combine with before for a window.",
"type": "string"
},
"deleteAll": {
"description": "Delete every statement but keep the graph: its name, URL, and settings (e.g. wikilinksMode) stay valid, so it can be rebuilt in place with create_knowledge_graph.",
"type": "boolean"
},
"statementIds": {
"description": "Advanced: delete by statement id (ids come from an earlier dry run of this tool or from retrieve_from_knowledge_base). Ids that do not belong to this graph are ignored and reported as ignoredIds.",
"type": "array",
"items": {
"type": "integer",
"minimum": -9007199254740991,
"maximum": 9007199254740991
}
},
"confirm": {
"default": false,
"description": "Without this (the default) the call is a DRY RUN that only reports what would be deleted. Set true, with the same filter, only after the user has agreed to delete what the dry run showed. The deletion is irreversible.",
"type": "boolean"
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"graphName",
"context"
]
}Annotations{
"readOnlyHint": false,
"idempotentHint": false,
"destructiveHint": true,
"openWorldHint": false
} | — | — |
develop_conceptual_bridgesAnalyze text or an existing graph and get ideas on how to develop conceptual bridges in this text to link it to a broader discourse. Provide either text, url, or graphName.Input schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"text": {
"description": "Text that you'd like to develop based on the latent concepts that connect this text to a broader discourse. Use new lines to separate separate statements or paragrams in each text (but not the sentences). Provide one of: text, url, or graphName.",
"type": "string"
},
"url": {
"description": "URL to fetch content from or YouTube video URL to fetch transcript. Provide one of: text, url, statements, or graphName.",
"type": "string",
"format": "uri"
},
"statements": {
"description": "Alternative to `text`: already-separated statements, each one unit of analysis (the role a line plays in `text`). Use only for short discrete items — notes, records, memory entries, survey answers, relations — especially when they carry `categories` or `timestamps`. For prose or long content use text/url instead, which is split on new lines anyway.",
"minItems": 1,
"type": "array",
"items": {
"type": "string",
"minLength": 1
}
},
"categories": {
"description": "Per-statement metadata labels, one entry per statement (empty array for none, length must match `statements`). Each label becomes a [[label]] node linked to that statement's concepts only, so author / source / tag / section can be filtered and grouped in the graph. Requires `statements`; omit when there is no metadata.",
"type": "array",
"items": {
"type": "array",
"items": {
"type": "string"
}
}
},
"timestamps": {
"description": "Per-statement date, one entry per statement (empty string = upload time, length must match `statements`). ISO 8601 only — '2026-08-02' or '2026-08-02T14:30:00Z'; other formats are refused because day-first and month-first dates are indistinguishable below the 13th. Stored to minute precision in the server timezone; drives time filters and dynamic graph views. Requires `statements`; omit when the statements are undated.",
"type": "array",
"items": {
"type": "string"
}
},
"graphName": {
"description": "Name of an existing InfraNodus graph to use. Provide one of: text, url, or graphName.",
"type": "string",
"minLength": 1
},
"requestMode": {
"default": "transcend",
"description": "Request mode: 'question' — generate questions that focus on this context; 'transcend' — generate responses that transcend and go beyond the content of the text and relate to a broader discourse.",
"type": "string",
"enum": [
"question",
"transcend"
]
},
"modelToUse": {
"default": "gpt-5.4",
"description": "AI model to use for generating research questions: claude-opus-5, claude-sonnet-5, claude-fable-5, gemini-2.5-pro, gemini-2.5-flash, gemini-2.5-flash-lite, grok-4.1-fast-non-reasoning, grok-4.1-fast-reasoning, gpt-4o, gpt-4o-mini, gpt-5.4, gpt-5.4-mini, gpt-5.6-terra, gpt-5.6-sol",
"type": "string",
"enum": [
"claude-opus-5",
"claude-sonnet-5",
"claude-fable-5",
"gemini-2.5-pro",
"gemini-2.5-flash",
"gemini-2.5-flash-lite",
"grok-4.1-fast-non-reasoning",
"grok-4.1-fast-reasoning",
"gpt-4o",
"gpt-4o-mini",
"gpt-5.4",
"gpt-5.4-mini",
"gpt-5.6-terra",
"gpt-5.6-sol"
]
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"context"
]
}Annotations{
"readOnlyHint": true,
"idempotentHint": true,
"destructiveHint": false
} | — | — |
develop_latent_topicsAnalyze text or an existing graph, extract underdeveloped topics and get an idea on how to develop them. Provide either text, url, or graphName.Input schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"text": {
"description": "Text that you'd like to develop based on the latent concepts that connect this text to a broader discourse. Use new lines to separate separate statements or paragrams in each text (but not the sentences). Provide one of: text, url, or graphName.",
"type": "string"
},
"url": {
"description": "URL to fetch content from or YouTube video URL to fetch transcript. Provide one of: text, url, statements, or graphName.",
"type": "string",
"format": "uri"
},
"statements": {
"description": "Alternative to `text`: already-separated statements, each one unit of analysis (the role a line plays in `text`). Use only for short discrete items — notes, records, memory entries, survey answers, relations — especially when they carry `categories` or `timestamps`. For prose or long content use text/url instead, which is split on new lines anyway.",
"minItems": 1,
"type": "array",
"items": {
"type": "string",
"minLength": 1
}
},
"categories": {
"description": "Per-statement metadata labels, one entry per statement (empty array for none, length must match `statements`). Each label becomes a [[label]] node linked to that statement's concepts only, so author / source / tag / section can be filtered and grouped in the graph. Requires `statements`; omit when there is no metadata.",
"type": "array",
"items": {
"type": "array",
"items": {
"type": "string"
}
}
},
"timestamps": {
"description": "Per-statement date, one entry per statement (empty string = upload time, length must match `statements`). ISO 8601 only — '2026-08-02' or '2026-08-02T14:30:00Z'; other formats are refused because day-first and month-first dates are indistinguishable below the 13th. Stored to minute precision in the server timezone; drives time filters and dynamic graph views. Requires `statements`; omit when the statements are undated.",
"type": "array",
"items": {
"type": "string"
}
},
"graphName": {
"description": "Name of an existing InfraNodus graph to use. Provide one of: text, url, or graphName.",
"type": "string",
"minLength": 1
},
"requestMode": {
"default": "transcend",
"description": "Request mode: 'question' — generate questions that focus on this context; 'transcend' — generate responses that transcend and go beyond the content of the text and relate to a broader discourse.",
"type": "string",
"enum": [
"question",
"transcend"
]
},
"modelToUse": {
"default": "gpt-5.4",
"description": "AI model to use for generating research questions: claude-opus-5, claude-sonnet-5, claude-fable-5, gemini-2.5-pro, gemini-2.5-flash, gemini-2.5-flash-lite, grok-4.1-fast-non-reasoning, grok-4.1-fast-reasoning, gpt-4o, gpt-4o-mini, gpt-5.4, gpt-5.4-mini, gpt-5.6-terra, gpt-5.6-sol",
"type": "string",
"enum": [
"claude-opus-5",
"claude-sonnet-5",
"claude-fable-5",
"gemini-2.5-pro",
"gemini-2.5-flash",
"gemini-2.5-flash-lite",
"grok-4.1-fast-non-reasoning",
"grok-4.1-fast-reasoning",
"gpt-4o",
"gpt-4o-mini",
"gpt-5.4",
"gpt-5.4-mini",
"gpt-5.6-terra",
"gpt-5.6-sol"
]
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"context"
]
}Annotations{
"readOnlyHint": true,
"idempotentHint": true,
"destructiveHint": false
} | — | — |
develop_text_toolAnalyze text or an existing graph to extract research questions, develop latent topics, and identify content gaps in a single workflow with progress tracking. Provide either text, url, or graphName.Input schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"text": {
"description": "Text that you'd like to think about and analyze. Use new lines to separate separate statements or paragrams in each text (but not the sentences). Provide one of: text, url, or graphName.",
"type": "string"
},
"url": {
"description": "URL to fetch content from or YouTube video URL to fetch transcript. Provide one of: text, url, statements, or graphName.",
"type": "string",
"format": "uri"
},
"statements": {
"description": "Alternative to `text`: already-separated statements, each one unit of analysis (the role a line plays in `text`). Use only for short discrete items — notes, records, memory entries, survey answers, relations — especially when they carry `categories` or `timestamps`. For prose or long content use text/url instead, which is split on new lines anyway.",
"minItems": 1,
"type": "array",
"items": {
"type": "string",
"minLength": 1
}
},
"categories": {
"description": "Per-statement metadata labels, one entry per statement (empty array for none, length must match `statements`). Each label becomes a [[label]] node linked to that statement's concepts only, so author / source / tag / section can be filtered and grouped in the graph. Requires `statements`; omit when there is no metadata.",
"type": "array",
"items": {
"type": "array",
"items": {
"type": "string"
}
}
},
"timestamps": {
"description": "Per-statement date, one entry per statement (empty string = upload time, length must match `statements`). ISO 8601 only — '2026-08-02' or '2026-08-02T14:30:00Z'; other formats are refused because day-first and month-first dates are indistinguishable below the 13th. Stored to minute precision in the server timezone; drives time filters and dynamic graph views. Requires `statements`; omit when the statements are undated.",
"type": "array",
"items": {
"type": "string"
}
},
"graphName": {
"description": "Name of an existing InfraNodus graph to use. Provide one of: text, url, or graphName.",
"type": "string",
"minLength": 1
},
"useSeveralGaps": {
"default": false,
"description": "Generate questions for several content gaps found in text",
"type": "boolean"
},
"gapDepth": {
"default": 0,
"description": "Depth of content gaps to generate questions for",
"type": "number"
},
"transcendDiscourse": {
"default": false,
"description": "Shall we transcend and go beyond this text to relate to a broader discourse? If false, the focus is on this text only.",
"type": "boolean"
},
"modelToUse": {
"default": "gpt-5.4",
"description": "AI model to use for generating insights: claude-opus-5, claude-sonnet-5, claude-fable-5, gemini-2.5-pro, gemini-2.5-flash, gemini-2.5-flash-lite, grok-4.1-fast-non-reasoning, grok-4.1-fast-reasoning, gpt-4o, gpt-4o-mini, gpt-5.4, gpt-5.4-mini, gpt-5.6-terra, gpt-5.6-sol",
"type": "string",
"enum": [
"claude-opus-5",
"claude-sonnet-5",
"claude-fable-5",
"gemini-2.5-pro",
"gemini-2.5-flash",
"gemini-2.5-flash-lite",
"grok-4.1-fast-non-reasoning",
"grok-4.1-fast-reasoning",
"gpt-4o",
"gpt-4o-mini",
"gpt-5.4",
"gpt-5.4-mini",
"gpt-5.6-terra",
"gpt-5.6-sol"
]
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"context"
]
}Annotations{
"readOnlyHint": true,
"idempotentHint": true,
"destructiveHint": false
} | — | — |
difference_between_textsExtract the conceptial relations that are missing in the first text, url, or InfraNodus graph but are present in the other textsInput schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"contexts": {
"minItems": 2,
"type": "array",
"items": {
"anyOf": [
{
"type": "object",
"properties": {
"text": {
"type": "string",
"minLength": 1,
"description": "Text content - use new lines to separate statements (but not sentences)."
}
},
"required": [
"text"
],
"additionalProperties": false,
"description": "Context from plain text."
},
{
"type": "object",
"properties": {
"statements": {
"minItems": 1,
"type": "array",
"items": {
"type": "string",
"minLength": 1
},
"description": "Short discrete statements, one unit of analysis each."
},
"categories": {
"description": "Per-statement metadata labels, one entry per statement (empty array for none, length must match `statements`). Each label becomes a [[label]] node linked to that statement's concepts only, so author / source / tag / section can be filtered and grouped in the graph. Requires `statements`; omit when there is no metadata.",
"type": "array",
"items": {
"type": "array",
"items": {
"type": "string"
}
}
},
"timestamps": {
"description": "Per-statement date, one entry per statement (empty string = upload time, length must match `statements`). ISO 8601 only — '2026-08-02' or '2026-08-02T14:30:00Z'; other formats are refused because day-first and month-first dates are indistinguishable below the 13th. Stored to minute precision in the server timezone; drives time filters and dynamic graph views. Requires `statements`; omit when the statements are undated.",
"type": "array",
"items": {
"type": "string"
}
}
},
"required": [
"statements"
],
"additionalProperties": false,
"description": "Context from short discrete statements. Categories and timestamps apply only when every context uses statements; otherwise the statements are joined into text."
},
{
"type": "object",
"properties": {
"url": {
"type": "string",
"minLength": 1,
"format": "uri",
"description": "URL to fetch content from (or YouTube transcript)."
}
},
"required": [
"url"
],
"additionalProperties": false,
"description": "Context from a URL."
},
{
"type": "object",
"properties": {
"graphName": {
"type": "string",
"minLength": 1,
"description": "Name of an existing InfraNodus graph; its statements are retrieved and used as text."
}
},
"required": [
"graphName"
],
"additionalProperties": false,
"description": "Context from an existing InfraNodus graph by name."
}
]
},
"description": "Array where the FIRST item is the target to analyze for missing parts; REMAINING items are reference sources. Each item is an object with exactly one of: { text: string }, { statements: string[] } (optionally with categories / timestamps), { url: string }, or { graphName: string }. Example: [{ text: '...' }, { url: 'https://...' }, { graphName: 'my-graph' }]."
},
"modifyAnalyzedText": {
"default": "none",
"description": "Entity detection: none (normal), detectEntities (mix entities and words), extractEntitiesOnly (detect entities only - use for ontology and knowledge graph creation and entity extraction)",
"type": "string",
"enum": [
"none",
"detectEntities",
"extractEntitiesOnly"
]
},
"includeStatements": {
"default": false,
"description": "Include processed statements in response (add only if explicitly needed)",
"type": "boolean"
},
"includeGraph": {
"default": false,
"description": "Include full graph structure in response (add only if explicitly needed)",
"type": "boolean"
},
"addNodesAndEdges": {
"default": false,
"description": "Include nodes and edges in response (add only if explicitly needed, not recommended for longer texts)",
"type": "boolean"
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"contexts",
"context"
]
}Annotations{
"readOnlyHint": true,
"idempotentHint": true,
"destructiveHint": false
} | — | — |
enable_project_learningsCreate the per-project learnings graph (learn-<slug>) in the user's InfraNodus account so that add_project_learnings can save what you learn about operating in this project. Call this ONLY when the user has explicitly asked to start saving learnings for a project — never on your own initiative. Before calling, tell the user in one or two sentences: what will be stored (knowledge about the project — where things live, traps, conventions, decisions, workflows — never anything about them personally, never secrets), where (a private, append-only graph in their InfraNodus account that they can delete at any time), and that each batch of learnings will be shown to them before it is saved unless they say otherwise. Idempotent: calling it again for an already enabled project just returns the graph.Input schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"project": {
"type": "string",
"minLength": 1,
"description": "Name of the project the learnings belong to — the repo or folder name, git remote basename, or whatever the user calls it. Use the same name every time for the same project (call get_project_learnings without a project to list the names already in use). Becomes the graph name learn-<slug>."
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"project",
"context"
]
}Annotations{
"readOnlyHint": false,
"idempotentHint": true,
"destructiveHint": false,
"openWorldHint": false
} | — | — |
fetchFetch a specific search result for an InfraNodus knowledge graphInput schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"id": {
"type": "string",
"minLength": 1,
"description": "ID of the search result to retrieve (username:graph_name:search_query"
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"id",
"context"
]
}Annotations{
"readOnlyHint": true,
"idempotentHint": true,
"destructiveHint": false
} | — | — |
generate_content_gapsGenerate content gaps from text, URL, or an existing graph using knowledge graph analysis.Input schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"text": {
"description": "Text that you'd like to retrieve content gaps from using knowledge graph analysis. Use new lines to separate separate statements or paragrams in each text (but not the sentences). Provide one of: text, url, or graphName.",
"type": "string"
},
"url": {
"description": "URL to fetch content from or YouTube video URL to fetch transcript. Provide one of: text, url, statements, or graphName.",
"type": "string",
"format": "uri"
},
"statements": {
"description": "Alternative to `text`: already-separated statements, each one unit of analysis (the role a line plays in `text`). Use only for short discrete items — notes, records, memory entries, survey answers, relations — especially when they carry `categories` or `timestamps`. For prose or long content use text/url instead, which is split on new lines anyway.",
"minItems": 1,
"type": "array",
"items": {
"type": "string",
"minLength": 1
}
},
"categories": {
"description": "Per-statement metadata labels, one entry per statement (empty array for none, length must match `statements`). Each label becomes a [[label]] node linked to that statement's concepts only, so author / source / tag / section can be filtered and grouped in the graph. Requires `statements`; omit when there is no metadata.",
"type": "array",
"items": {
"type": "array",
"items": {
"type": "string"
}
}
},
"timestamps": {
"description": "Per-statement date, one entry per statement (empty string = upload time, length must match `statements`). ISO 8601 only — '2026-08-02' or '2026-08-02T14:30:00Z'; other formats are refused because day-first and month-first dates are indistinguishable below the 13th. Stored to minute precision in the server timezone; drives time filters and dynamic graph views. Requires `statements`; omit when the statements are undated.",
"type": "array",
"items": {
"type": "string"
}
},
"graphName": {
"description": "Name of an existing InfraNodus graph to use. Provide one of: text, url, or graphName.",
"type": "string",
"minLength": 1
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"context"
]
}Annotations{
"readOnlyHint": true,
"idempotentHint": true,
"destructiveHint": false
} | — | — |
generate_contextual_hintGenerate information about the main topics and concepts in a text to augment RAG retrieval and text analysis.Input schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"text": {
"description": "Text that you'd like to get an overview of to augment RAG retrieval and text analysis. Use new lines to separate separate statements or paragrams in each text (but not the sentences). Provide one of: text, url, or graphName.",
"type": "string"
},
"url": {
"description": "URL to fetch content from or YouTube video URL to fetch transcript. Provide one of: text, url, statements, or graphName.",
"type": "string",
"format": "uri"
},
"statements": {
"description": "Alternative to `text`: already-separated statements, each one unit of analysis (the role a line plays in `text`). Use only for short discrete items — notes, records, memory entries, survey answers, relations — especially when they carry `categories` or `timestamps`. For prose or long content use text/url instead, which is split on new lines anyway.",
"minItems": 1,
"type": "array",
"items": {
"type": "string",
"minLength": 1
}
},
"categories": {
"description": "Per-statement metadata labels, one entry per statement (empty array for none, length must match `statements`). Each label becomes a [[label]] node linked to that statement's concepts only, so author / source / tag / section can be filtered and grouped in the graph. Requires `statements`; omit when there is no metadata.",
"type": "array",
"items": {
"type": "array",
"items": {
"type": "string"
}
}
},
"timestamps": {
"description": "Per-statement date, one entry per statement (empty string = upload time, length must match `statements`). ISO 8601 only — '2026-08-02' or '2026-08-02T14:30:00Z'; other formats are refused because day-first and month-first dates are indistinguishable below the 13th. Stored to minute precision in the server timezone; drives time filters and dynamic graph views. Requires `statements`; omit when the statements are undated.",
"type": "array",
"items": {
"type": "string"
}
},
"graphName": {
"description": "Name of an existing InfraNodus graph to use. Provide one of: text, url, or graphName.",
"type": "string",
"minLength": 1
},
"userName": {
"description": "Name of the user who created the public graph. Provide only if username is explicitly requested by the user and their public graph is requested, otherwise do not provide.",
"type": "string"
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"context"
]
}Annotations{
"readOnlyHint": true,
"idempotentHint": true,
"destructiveHint": false
} | — | — |
generate_knowledge_graphGenerate a knowledge graph with main topics, topical clusters, concepts, concepts (nodes) relations (edges) and structural gaps. Only use when explicitly asked to analyze a text or generate a knowledge graph. Do not use for short clarifying questions that you already have an answer to from the context of the conversation.Input schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"text": {
"description": "Text that you'd like to analyze. Use new lines to separate separate statements or paragrams in each text (but not the sentences). Use [[wikilinks]] to mark entities (if required for social / knowledge graphs, ontology, or entity detection). Provide one of: this, url, or statements.",
"type": "string"
},
"url": {
"description": "URL to fetch content from or YouTube video URL to fetch transcript. Provide either this or text, not both.",
"type": "string",
"format": "uri"
},
"statements": {
"description": "Alternative to `text`: already-separated statements, each one unit of analysis (the role a line plays in `text`). Use only for short discrete items — notes, records, memory entries, survey answers, relations — especially when they carry `categories` or `timestamps`. For prose or long content use text/url instead, which is split on new lines anyway.",
"minItems": 1,
"type": "array",
"items": {
"type": "string",
"minLength": 1
}
},
"categories": {
"description": "Per-statement metadata labels, one entry per statement (empty array for none, length must match `statements`). Each label becomes a [[label]] node linked to that statement's concepts only, so author / source / tag / section can be filtered and grouped in the graph. Requires `statements`; omit when there is no metadata.",
"type": "array",
"items": {
"type": "array",
"items": {
"type": "string"
}
}
},
"timestamps": {
"description": "Per-statement date, one entry per statement (empty string = upload time, length must match `statements`). ISO 8601 only — '2026-08-02' or '2026-08-02T14:30:00Z'; other formats are refused because day-first and month-first dates are indistinguishable below the 13th. Stored to minute precision in the server timezone; drives time filters and dynamic graph views. Requires `statements`; omit when the statements are undated.",
"type": "array",
"items": {
"type": "string"
}
},
"includeStatements": {
"default": false,
"description": "Include processed statements in response (true only if explicitly needed or requested)",
"type": "boolean"
},
"includeGraph": {
"default": true,
"description": "Include full graph structure in response (true only if explicitly needed or requested)",
"type": "boolean"
},
"addNodesAndEdges": {
"default": false,
"description": "Include nodes and edges in response (true only if explicitly needed, not recommended for longer texts)",
"type": "boolean"
},
"fullGraph": {
"default": false,
"description": "Return the complete non-compacted graph: all node attributes (degree, betweenness centrality, community), all edge attributes including context_matrix (which statements produced each edge, with weights), the nodes-to-statements map, and full statement metadata. Overrides includeGraph / addNodesAndEdges / compaction. Token-heavy — use only when the raw graph data is explicitly needed (e.g. export, rendering, or programmatic processing).",
"type": "boolean"
},
"maxNodes": {
"description": "Maximum number of concepts (nodes) in the generated graph. Default 150 — the most relevant nodes are kept and the rest dropped. Increase (e.g. 500) for large texts when a more complete graph is explicitly needed; response size grows accordingly. Omit to use the default.",
"type": "integer",
"exclusiveMinimum": 0,
"maximum": 1000
},
"modifyAnalyzedText": {
"default": "none",
"description": "Text processing setting to use: none (for text, gap, and topical analysis), detectEntities (mix entities and words), extractEntitiesOnly (detect entities only - use for ontology and knowledge graph generation and entity extraction)",
"type": "string",
"enum": [
"none",
"detectEntities",
"extractEntitiesOnly"
]
},
"wikilinksMode": {
"default": "default",
"description": "How [[wikilinks]] in the text become graph nodes. 'default': wikilinks become [[entity]] nodes; statements that contain wikilinks use only those, statements without any are processed word-by-word. 'wikilinksOnly': ONLY [[wikilinks]] become nodes, all other words are ignored; nearby wikilinks connect via a sliding co-occurrence window — use for entity/ontology/link statements like '[[alice]] works with [[bob]]' or '[[Page A]] links to [[Page B]]'. 'obsidianStyle': like Obsidian's graph view — markdown headings set the parent page for the statements below them ('## [[Page A]]' or a plain '## Section Title'), and a line may override its parent with a '[[Parent Page]]: ' prefix (colon required); the parent becomes a [[page]] node connected to every concept of its statements, while concepts do NOT connect to each other (star topology around pages/sections). 'parentAndConcepts': same heading/prefix parent contract, but concepts also keep their normal co-occurrence connections — best for mixed content (notes, docs, articles) where you want both section/page provenance and a real concept graph. 'plainText': brackets are stripped and everything is processed as ordinary words. In all modes parent and wikilink nodes share the [[name]] namespace, so graphs remain comparable/mergeable. With `statements` + `categories` supplied directly, parent extraction is skipped and those categories become the mention labels ('obsidianStyle' still gives the star topology). Takes effect when the graph is generated or first created; an existing saved graph keeps its original setting.",
"type": "string",
"enum": [
"default",
"wikilinksOnly",
"obsidianStyle",
"parentAndConcepts",
"plainText"
]
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"context"
]
}Annotations{
"readOnlyHint": true,
"idempotentHint": true,
"destructiveHint": false
} | — | — |
generate_ontology_graphUse AI to generate a reasoning ontology knowledge graph (entities and the relations between them) and optionally save it as a InfraNodus graph. Three sources, provide exactly one: prompt (a topic — one AI call), text (a long document or a structural digest of a project, chunked server-side), or sourceGraphName (an existing graph — e.g. a fully ingested repo, vault, or corpus — whose statements are read back, chunked, and condensed into an ontology). Set ontologyMode: 'codebase' for software projects, or 'procedural' to write a DIGEST of how the project works (prose statements with [[wikilinks]], not relation triples) from an already-uploaded graph — save it as <repo|vault>-<project>-digest for optimize_knowledge_base. Use to get a rich overview, a reasoning map of a topic, or a condensed 'how it fits together' graph of a large corpus.Input schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"prompt": {
"description": "A topic or short prompt to generate a reasoning ontology for, e.g. 'build an ontology on AI attention mechanisms' or 'the main principles of Ray Dalio applied to investment'. Sent to the AI as-is in one call. Provide exactly one of: prompt, text, sourceGraphName.",
"type": "string"
},
"text": {
"description": "A long text, document, or structure map of a project (file tree, imports, exports, docstring headlines) to EXTRACT an ontology from. Chunked server-side (see chunkSize); every chunk's ontology is appended to the same graph. Provide exactly one of: prompt, text, sourceGraphName.",
"type": "string"
},
"sourceGraphName": {
"description": "Name of an existing InfraNodus graph whose statements are the source: they are read back, chunked, and an ontology is generated from each chunk into the graph named by graphName. Use it to condense a fully ingested repo, vault, or corpus graph (e.g. repo-<project>-docs or repo-<project>-structure) into an ontology, or — with ontologyMode 'procedural' — into a digest of how the project works. Provide exactly one of: prompt, text, sourceGraphName.",
"type": "string"
},
"ontologyMode": {
"default": "general",
"description": "'general' (default): entities and relations of any domain. 'codebase': the source is a software project's structure map or code documentation — entities are modules/files, functions, classes, data stores, external services, configuration, and the domain concepts they implement; relations favour dependsOn / partOf / exposes / calls / stores / implements. 'procedural': instead of an ontology, write a DIGEST of how the project works — one prose statement per line with [[wikilinks]] on the modules, concepts, and files involved, each ending with its type in [brackets] — [principles], [rules], [procedures], [handoffs], [main_ideas], [gaps] — which is saved as the statement's category; save it as repo-<project>-digest or vault-<project>-digest and feed it to optimize_knowledge_base. Use it when the content already lives in a graph (docs, structure, notes) and nobody can or should re-read the files. Applies to text and sourceGraphName inputs.",
"type": "string",
"enum": [
"general",
"codebase",
"procedural"
]
},
"chunkSize": {
"default": 12000,
"description": "Characters per chunk for text / sourceGraphName inputs (split on line boundaries). Smaller chunks give more detailed, more expensive ontologies. Default 12000.",
"type": "integer",
"minimum": 2000,
"maximum": 40000
},
"graphName": {
"description": "Name of the InfraNodus graph to save the ontology to. Only used when saveGraph is true. If omitted, a name is auto-generated from the prompt. Note: saving to a name that already exists appends the new statements to that graph rather than replacing it.",
"type": "string"
},
"modelToUse": {
"default": "claude-opus-5",
"description": "AI model used to generate the ontology. More capable models (claude-opus-5, claude-fable-5, gpt-5.6-terra) produce richer, more accurate ontologies; the -mini and -lite variants are faster and cheaper. Default: claude-opus-5.",
"type": "string",
"enum": [
"claude-opus-5",
"claude-sonnet-5",
"claude-fable-5",
"gemini-2.5-pro",
"gemini-2.5-flash",
"gemini-2.5-flash-lite",
"grok-4.1-fast-non-reasoning",
"grok-4.1-fast-reasoning",
"gpt-4o",
"gpt-4o-mini",
"gpt-5.4",
"gpt-5.4-mini",
"gpt-5.6-terra",
"gpt-5.6-sol"
]
},
"saveGraph": {
"default": true,
"description": "Whether to save the generated ontology as a persistent InfraNodus graph (true by default). When true, the graph is stored under your account and a link is returned. Set to false if the user explicitly asks not to save, or when you only need a one-off AI ontology overview of a topic for the current context that won't be reused later — in that case the generated ontology statements are returned directly without being persisted.",
"type": "boolean"
},
"includeGraph": {
"default": false,
"description": "Include the compact graph structure — nodes (entities), edges (relations between them), and clusters — in the response. False by default to keep the response small; the ontology statements and analytics usually carry what's needed. Set to true when you also want to inspect the node/edge structure or render it.",
"type": "boolean"
},
"fullGraph": {
"default": false,
"description": "Return the complete non-compacted graph: all node attributes (degree, betweenness centrality, community), all edge attributes including context_matrix (which statements produced each edge, with weights), and the nodes-to-statements map. Implies includeGraph and overrides compaction. Token-heavy — use only when the raw graph data is explicitly needed (e.g. export, rendering, or programmatic processing).",
"type": "boolean"
},
"maxNodes": {
"description": "Maximum number of concepts (nodes) in the generated graph. Default 150 — the most relevant nodes are kept and the rest dropped. Increase (e.g. 500) for large ontologies when a more complete graph is explicitly needed; response size grows accordingly. Omit to use the default.",
"type": "integer",
"exclusiveMinimum": 0,
"maximum": 1000
},
"includeAnalytics": {
"default": true,
"description": "Include graph analytics — main topical clusters, content gaps, top influential nodes / concepts, top relations, conceptual gateways, and network statistics — derived from the generated ontology. True by default. Use this to get insights from the graph: keep it on when you need to understand the structure, gaps, or key concepts; turn off to save context space when you only need the raw ontology statements (see includeStatements) or the graph link.",
"type": "boolean"
},
"includeStatements": {
"default": true,
"description": "Include the ontology statements — the entity-relation statements the AI generated, in their final post-processing shape as stored in the graph — in the response (the ontologyStatements array). True by default so you can read the generated ontology. Set to false to save context space when you only need the analytics and/or the saved graph link, not the underlying statements.",
"type": "boolean"
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"context"
]
}Annotations{
"readOnlyHint": false,
"idempotentHint": false,
"destructiveHint": false
} | — | — |
generate_research_ideasAnalyze text or an existing graph and generate innovative research ideas based on the content gaps identified between the topical clusters inside the text that can be used to improve the text and the discourse it relates to.Input schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"text": {
"description": "Text that you'd like to generate research ideas from based on the content gaps identified between the topical clusters inside the text. Use new lines to separate separate statements or paragrams in each text (but not the sentences). Provide one of: text, url, or graphName.",
"type": "string"
},
"url": {
"description": "URL to fetch content from or YouTube video URL to fetch transcript. Provide one of: text, url, statements, or graphName.",
"type": "string",
"format": "uri"
},
"statements": {
"description": "Alternative to `text`: already-separated statements, each one unit of analysis (the role a line plays in `text`). Use only for short discrete items — notes, records, memory entries, survey answers, relations — especially when they carry `categories` or `timestamps`. For prose or long content use text/url instead, which is split on new lines anyway.",
"minItems": 1,
"type": "array",
"items": {
"type": "string",
"minLength": 1
}
},
"categories": {
"description": "Per-statement metadata labels, one entry per statement (empty array for none, length must match `statements`). Each label becomes a [[label]] node linked to that statement's concepts only, so author / source / tag / section can be filtered and grouped in the graph. Requires `statements`; omit when there is no metadata.",
"type": "array",
"items": {
"type": "array",
"items": {
"type": "string"
}
}
},
"timestamps": {
"description": "Per-statement date, one entry per statement (empty string = upload time, length must match `statements`). ISO 8601 only — '2026-08-02' or '2026-08-02T14:30:00Z'; other formats are refused because day-first and month-first dates are indistinguishable below the 13th. Stored to minute precision in the server timezone; drives time filters and dynamic graph views. Requires `statements`; omit when the statements are undated.",
"type": "array",
"items": {
"type": "string"
}
},
"graphName": {
"description": "Name of an existing InfraNodus graph to use. Provide one of: text, url, or graphName.",
"type": "string",
"minLength": 1
},
"userName": {
"description": "Name of the user who created the public graph. Provide only if username is explicitly requested by the user and their public graph is requested, otherwise do not provide.",
"type": "string"
},
"useSeveralGaps": {
"default": false,
"description": "Generate ideas for several content gaps found in text",
"type": "boolean"
},
"gapDepth": {
"default": 0,
"description": "Depth of content gaps to generate ideas for",
"type": "number"
},
"shouldTranscend": {
"default": false,
"description": "Generate ideas that transcend and go beyond the content of the text and relate to a broader discourse. Only run if explicitly requested by the user to go beyond the text and relate to a broader discourse",
"type": "boolean"
},
"responseType": {
"default": "response",
"description": "Type of response to generate: 'response' — generates a response based on the gaps identified; 'idea' — generate an business idea that bridges the gap.",
"type": "string",
"enum": [
"response",
"idea"
]
},
"includeGraphSummary": {
"default": false,
"description": "Include graph summary string in the response to provide additional information about the text knowledge graph structure",
"type": "boolean"
},
"modelToUse": {
"default": "gpt-5.4",
"description": "AI model to use for generating research questions: claude-opus-5, claude-sonnet-5, claude-fable-5, gemini-2.5-pro, gemini-2.5-flash, gemini-2.5-flash-lite, grok-4.1-fast-non-reasoning, grok-4.1-fast-reasoning, gpt-4o, gpt-4o-mini, gpt-5.4, gpt-5.4-mini, gpt-5.6-terra, gpt-5.6-sol",
"type": "string",
"enum": [
"claude-opus-5",
"claude-sonnet-5",
"claude-fable-5",
"gemini-2.5-pro",
"gemini-2.5-flash",
"gemini-2.5-flash-lite",
"grok-4.1-fast-non-reasoning",
"grok-4.1-fast-reasoning",
"gpt-4o",
"gpt-4o-mini",
"gpt-5.4",
"gpt-5.4-mini",
"gpt-5.6-terra",
"gpt-5.6-sol"
]
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"context"
]
}Annotations{
"readOnlyHint": true,
"idempotentHint": true,
"destructiveHint": false
} | — | — |
generate_research_questionsAnalyze text or an existing graph and generate innovative research questions based on the content gaps identified between the topical clusters. Provide either text, url, or graphName. Can be used to improve the text and the discourse it relates toInput schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"text": {
"description": "Text that you'd like to generate research questions from. Use new lines to separate separate statements or paragrams in each text (but not the sentences). Provide one of: text, url, or graphName.",
"type": "string"
},
"url": {
"description": "URL to fetch content from or YouTube video URL to fetch transcript. Provide one of: text, url, statements, or graphName.",
"type": "string",
"format": "uri"
},
"statements": {
"description": "Alternative to `text`: already-separated statements, each one unit of analysis (the role a line plays in `text`). Use only for short discrete items — notes, records, memory entries, survey answers, relations — especially when they carry `categories` or `timestamps`. For prose or long content use text/url instead, which is split on new lines anyway.",
"minItems": 1,
"type": "array",
"items": {
"type": "string",
"minLength": 1
}
},
"categories": {
"description": "Per-statement metadata labels, one entry per statement (empty array for none, length must match `statements`). Each label becomes a [[label]] node linked to that statement's concepts only, so author / source / tag / section can be filtered and grouped in the graph. Requires `statements`; omit when there is no metadata.",
"type": "array",
"items": {
"type": "array",
"items": {
"type": "string"
}
}
},
"timestamps": {
"description": "Per-statement date, one entry per statement (empty string = upload time, length must match `statements`). ISO 8601 only — '2026-08-02' or '2026-08-02T14:30:00Z'; other formats are refused because day-first and month-first dates are indistinguishable below the 13th. Stored to minute precision in the server timezone; drives time filters and dynamic graph views. Requires `statements`; omit when the statements are undated.",
"type": "array",
"items": {
"type": "string"
}
},
"graphName": {
"description": "Name of an existing InfraNodus graph in your account to generate research questions from. Provide one of: text, url, or graphName.",
"type": "string",
"minLength": 1
},
"userName": {
"description": "Name of the user who created the public graph. Provide only if username is explicitly requested by the user and their public graph is requested, otherwise do not provide.",
"type": "string"
},
"useSeveralGaps": {
"default": false,
"description": "Generate questions for several content gaps found in text",
"type": "boolean"
},
"gapDepth": {
"default": 0,
"description": "Depth of content gaps to generate questions for",
"type": "number"
},
"includeGraphSummary": {
"default": false,
"description": "Include graph summary string in the response to provide additional information about the text knowledge graph structure",
"type": "boolean"
},
"modelToUse": {
"default": "gpt-5.4",
"description": "AI model to use for generating research questions: claude-opus-5, claude-sonnet-5, claude-fable-5, gemini-2.5-pro, gemini-2.5-flash, gemini-2.5-flash-lite, grok-4.1-fast-non-reasoning, grok-4.1-fast-reasoning, gpt-4o, gpt-4o-mini, gpt-5.4, gpt-5.4-mini, gpt-5.6-terra, gpt-5.6-sol",
"type": "string",
"enum": [
"claude-opus-5",
"claude-sonnet-5",
"claude-fable-5",
"gemini-2.5-pro",
"gemini-2.5-flash",
"gemini-2.5-flash-lite",
"grok-4.1-fast-non-reasoning",
"grok-4.1-fast-reasoning",
"gpt-4o",
"gpt-4o-mini",
"gpt-5.4",
"gpt-5.4-mini",
"gpt-5.6-terra",
"gpt-5.6-sol"
]
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"context"
]
}Annotations{
"readOnlyHint": true,
"idempotentHint": true,
"destructiveHint": false
} | — | — |
generate_responses_from_graphUse text, URL, or an existing InfraNodus knowledge graph and generate responses and expert advice based on a prompt provided.Input schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"graphName": {
"description": "Name of an existing InfraNodus graph in your account to retrieve. Provide one of: text, url, or graphName.",
"type": "string",
"minLength": 1
},
"userName": {
"description": "Name of the user who created the public graph. Provide only if username is explicitly requested by the user and their public graph is requested, otherwise do not provide.",
"type": "string"
},
"text": {
"description": "Text that you'd like to generate responses from. Use new lines to separate separate statements or paragrams in each text (but not the sentences). Provide one of: text, url, or graphName.",
"type": "string"
},
"url": {
"description": "URL to fetch content from or YouTube video URL to fetch transcript. Provide one of: text, url, statements, or graphName.",
"type": "string",
"format": "uri"
},
"statements": {
"description": "Alternative to `text`: already-separated statements, each one unit of analysis (the role a line plays in `text`). Use only for short discrete items — notes, records, memory entries, survey answers, relations — especially when they carry `categories` or `timestamps`. For prose or long content use text/url instead, which is split on new lines anyway.",
"minItems": 1,
"type": "array",
"items": {
"type": "string",
"minLength": 1
}
},
"categories": {
"description": "Per-statement metadata labels, one entry per statement (empty array for none, length must match `statements`). Each label becomes a [[label]] node linked to that statement's concepts only, so author / source / tag / section can be filtered and grouped in the graph. Requires `statements`; omit when there is no metadata.",
"type": "array",
"items": {
"type": "array",
"items": {
"type": "string"
}
}
},
"timestamps": {
"description": "Per-statement date, one entry per statement (empty string = upload time, length must match `statements`). ISO 8601 only — '2026-08-02' or '2026-08-02T14:30:00Z'; other formats are refused because day-first and month-first dates are indistinguishable below the 13th. Stored to minute precision in the server timezone; drives time filters and dynamic graph views. Requires `statements`; omit when the statements are undated.",
"type": "array",
"items": {
"type": "string"
}
},
"prompt": {
"type": "string",
"minLength": 1,
"description": "Prompt to generate responses to from the graph"
},
"modelToUse": {
"default": "gpt-5.4",
"description": "AI model to use for generating research questions: claude-opus-5, claude-sonnet-5, claude-fable-5, gemini-2.5-pro, gemini-2.5-flash, gemini-2.5-flash-lite, grok-4.1-fast-non-reasoning, grok-4.1-fast-reasoning, gpt-4o, gpt-4o-mini, gpt-5.4, gpt-5.4-mini, gpt-5.6-terra, gpt-5.6-sol",
"type": "string",
"enum": [
"claude-opus-5",
"claude-sonnet-5",
"claude-fable-5",
"gemini-2.5-pro",
"gemini-2.5-flash",
"gemini-2.5-flash-lite",
"grok-4.1-fast-non-reasoning",
"grok-4.1-fast-reasoning",
"gpt-4o",
"gpt-4o-mini",
"gpt-5.4",
"gpt-5.4-mini",
"gpt-5.6-terra",
"gpt-5.6-sol"
]
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"prompt",
"context"
]
}Annotations{
"readOnlyHint": true,
"idempotentHint": true,
"destructiveHint": false
} | — | — |
generate_seo_reportAnalyze content for SEO optimization by comparing its knowledge graph with the graphs of Google search results and search queries to identify content gaps and opportunities based on the differencesInput schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"text": {
"description": "Content that you'd like to optimize for SEO. Use new lines to separate separate statements or paragrams in each text (but not the sentences).",
"type": "string",
"minLength": 1
},
"url": {
"description": "URL to fetch content from for SEO analysis. Provide either this or text, not both.",
"type": "string",
"format": "uri"
},
"statements": {
"description": "Alternative to `text`: already-separated statements, each one unit of analysis (the role a line plays in `text`). Use only for short discrete items — notes, records, memory entries, survey answers, relations — especially when they carry `categories` or `timestamps`. For prose or long content use text/url instead, which is split on new lines anyway.",
"minItems": 1,
"type": "array",
"items": {
"type": "string",
"minLength": 1
}
},
"categories": {
"description": "Per-statement metadata labels, one entry per statement (empty array for none, length must match `statements`). Each label becomes a [[label]] node linked to that statement's concepts only, so author / source / tag / section can be filtered and grouped in the graph. Requires `statements`; omit when there is no metadata.",
"type": "array",
"items": {
"type": "array",
"items": {
"type": "string"
}
}
},
"contentToExtract": {
"default": "all",
"description": "What to extract from URL: 'all' (default), 'header tags', or 'link tags'.",
"type": "string",
"enum": [
"all",
"header tags",
"link tags"
]
},
"numberOfKeywordsToExtract": {
"default": 2,
"description": "Number of the top keyword groups extracted from text to use for the SEO analysis, default is 2, maximum is 4. Use more if you want to get deeper insights but longer processing. In case the tool fails, reduce to 2 or less.",
"type": "number",
"maximum": 4
},
"numberOfTopicsToExtract": {
"default": 2,
"description": "Number of the top topical cluster names extracted from text to use for the SEO analysis, default is 2, maximum is 4. Use more if you want to get deeper insights but longer processing. In case the tool fails, reduce to 2 or less.",
"type": "number",
"maximum": 4
},
"importLanguage": {
"default": "EN",
"description": "Language of the content and search queries, default is English (EN), use the language of the conversation or requested by user.",
"type": "string",
"enum": [
"EN",
"DE",
"FR",
"ES",
"IT",
"PT",
"RU",
"CN",
"JP",
"NL",
"TW",
"KO",
"AR",
"HE"
]
},
"importCountry": {
"default": "US",
"description": "Country for the search analysis, default is United States (US). Use the country most suitable for the language selected.",
"type": "string",
"enum": [
"AR",
"AU",
"BR",
"CA",
"CH",
"CN",
"DE",
"ES",
"FR",
"GB",
"HK",
"IN",
"IT",
"JP",
"MX",
"NL",
"NO",
"NZ",
"PT",
"RU",
"SV",
"TW",
"US",
"SG",
"SA",
"AE",
"EG",
"IR",
"IL",
"KR"
]
},
"useProxy": {
"default": false,
"description": "Use proxy to fetch content from URL (true only if the first request fails, returns javascript only,or is requested by user)",
"type": "boolean"
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"context"
]
}Annotations{
"readOnlyHint": true,
"idempotentHint": true,
"destructiveHint": false
} | — | — |
generate_topical_clustersGenerate topics and clusters of keywords from text, URL, or an existing graph using knowledge graph analysis. Input schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"text": {
"description": "Text that you'd like to retrieve topics and topical clusters from using knowledge graph analysis. Use new lines to separate separate statements or paragrams in each text (but not the sentences). Provide one of: text, url, or graphName.",
"type": "string"
},
"url": {
"description": "URL to fetch content from or YouTube video URL to fetch transcript. Provide one of: text, url, statements, or graphName.",
"type": "string",
"format": "uri"
},
"statements": {
"description": "Alternative to `text`: already-separated statements, each one unit of analysis (the role a line plays in `text`). Use only for short discrete items — notes, records, memory entries, survey answers, relations — especially when they carry `categories` or `timestamps`. For prose or long content use text/url instead, which is split on new lines anyway.",
"minItems": 1,
"type": "array",
"items": {
"type": "string",
"minLength": 1
}
},
"categories": {
"description": "Per-statement metadata labels, one entry per statement (empty array for none, length must match `statements`). Each label becomes a [[label]] node linked to that statement's concepts only, so author / source / tag / section can be filtered and grouped in the graph. Requires `statements`; omit when there is no metadata.",
"type": "array",
"items": {
"type": "array",
"items": {
"type": "string"
}
}
},
"timestamps": {
"description": "Per-statement date, one entry per statement (empty string = upload time, length must match `statements`). ISO 8601 only — '2026-08-02' or '2026-08-02T14:30:00Z'; other formats are refused because day-first and month-first dates are indistinguishable below the 13th. Stored to minute precision in the server timezone; drives time filters and dynamic graph views. Requires `statements`; omit when the statements are undated.",
"type": "array",
"items": {
"type": "string"
}
},
"graphName": {
"description": "Name of an existing InfraNodus graph to use. Provide one of: text, url, or graphName.",
"type": "string",
"minLength": 1
},
"userName": {
"description": "Name of the user who created the public graph. Provide only if username is explicitly requested by the user and their public graph is requested, otherwise do not provide.",
"type": "string"
},
"generateTopicalSummaries": {
"default": true,
"description": "Generates AI-powered summaries of each topical cluster (returned as topicalClusterSummaries), summarizing the discourse each cluster represents — useful for SEO-optimized content creation. True by default. Set to false to increase processing speed or if the summary request fails.",
"type": "boolean"
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"context"
]
}Annotations{
"readOnlyHint": true,
"idempotentHint": true,
"destructiveHint": false
} | — | — |
get_more_toolsCheck for additional tools whenever your task might benefit from specialized capabilities - even if existing tools could work as a fallback.Input schema{
"type": "object",
"properties": {
"context": {
"type": "string",
"description": "A description of your goal and what kind of tool would help accomplish it."
}
},
"required": [
"context"
]
} | — | — |
get_project_learningsRetrieve what previous sessions learned about operating in a project (saved with add_project_learnings). Call it at the start of a substantive task with the task as `prompt` to get the most relevant learnings plus an overview of what is known; with `entity` (a file path, module, or concept) before working on an unfamiliar area; with neither for a structural overview; or with no `project` to list the projects that have learnings in this account. Returns enabled: false with an empty list when the project has no learnings graph — that is not an error, just carry on; do not suggest enabling unless the user asks about memory.Input schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"project": {
"description": "Project name (same as used when enabling). Omit to list the projects that have learnings enabled in this account.",
"type": "string"
},
"prompt": {
"description": "The task or question at hand. Retrieves the most relevant learnings for it (GraphRAG) plus a structural overview of what is known. Use at the start of a substantive task.",
"type": "string"
},
"entity": {
"description": "A file path, module, or concept. Returns every learning that mentions it. Use before working on an unfamiliar area.",
"type": "string"
},
"type": {
"description": "Only return learnings of this type (e.g. 'trap' for the fragile areas of the project, 'approach' for what worked well or should be done differently). Applies to the overview and prompt modes.",
"type": "string",
"enum": [
"location",
"trap",
"convention",
"decision",
"workflow",
"question",
"approach"
]
},
"limit": {
"default": 15,
"description": "Maximum number of learnings to return.",
"type": "integer",
"minimum": 1,
"maximum": 50
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"context"
]
}Annotations{
"readOnlyHint": true,
"idempotentHint": true,
"destructiveHint": false,
"openWorldHint": false
} | — | — |
list_graphsList all graphs (contexts) for the currently logged in user with optional filtering by name, type, date, language, or favorite status. Use this to discover available graphs before analyzing or searching them.Input schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"nameContains": {
"description": "Values that should be matched to in the graph name. Use comma-separated values for OR logic (e.g., 'youtube,google,evernote'). Leave empty to list all graphs.",
"type": "string"
},
"userName": {
"description": "Show public graphs belonging to a certain user. Only provide if explicitly requested by the user.",
"type": "string"
},
"type": {
"description": "Filter by graph type. Available types: STANDARD, MINDMAP, WORDCLOUD, GEXF, SCIENCE, TWITTER, GOOGLE, NICHE, SEO, EVERNOTE, RSS, WIKILINKS, TXT, PDF, CSV, MD, MEMORY, ONTOLOGY, JSON, KWRDS. Use comma-separated values for OR logic (e.g., 'CSV,GOOGLE,STANDARD'). When user asks for a memory use, MEMORY in the type, when user asks for ontology, use ONTOLOGY,WIKILINKS. If nothing found or unsure what to use,leave empty.",
"type": "string"
},
"fromDate": {
"description": "Filter graphs created on or after this date (ISO format, e.g., '2026-01-01T00:00:00.000Z')",
"type": "string"
},
"toDate": {
"description": "Filter graphs created on or before this date (ISO format, e.g., '2026-01-31T23:59:59.999Z')",
"type": "string"
},
"language": {
"description": "Filter by language code (e.g., 'EN', 'AUTO', 'DE', 'FR', 'ES', etc.)",
"type": "string"
},
"favorite": {
"description": "Filter by favorite status",
"type": "boolean"
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"context"
]
}Annotations{
"readOnlyHint": true,
"idempotentHint": true,
"destructiveHint": false
} | — | — |
memory_add_relationsAdd relations to the InfraNodus memory from text, save it, and provide its name and a link to it for future use. Input schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"graphName": {
"type": "string",
"minLength": 1,
"maxLength": 28,
"description": "Name of the graph to add the memory to in InfraNodus - lowercase, dashes for spaces, no special characters. Auto-generate from the context of the conversation (if previously available) or use the nanme of the LLM client or project, or use the name the user explicitly provided or requested."
},
"text": {
"description": "Text that you'd like to analyze. Use new lines to separate separate statements, relations, and paragraphs in each text (but not the sentences). Detect the entities in every statement and use [[wikilinks]] syntax to mark them, unless the user explicitly requests automatic entity detection. Every statement should have at least two entities marked. Provide either this or statements.",
"type": "string"
},
"statements": {
"description": "Alternative to `text`: already-separated statements, each one unit of analysis (the role a line plays in `text`). Use only for short discrete items — notes, records, memory entries, survey answers, relations — especially when they carry `categories` or `timestamps`. For prose or long content use text/url instead, which is split on new lines anyway.",
"minItems": 1,
"type": "array",
"items": {
"type": "string",
"minLength": 1
}
},
"categories": {
"description": "Per-statement metadata labels, one entry per statement (empty array for none, length must match `statements`). Each label becomes a [[label]] node linked to that statement's concepts only, so author / source / tag / section can be filtered and grouped in the graph. Takes effect only when the graph is FIRST created — uploads to an existing graphName keep its original settings and the labels are ignored. Requires `statements`; omit when there is no metadata.",
"type": "array",
"items": {
"type": "array",
"items": {
"type": "string"
}
}
},
"timestamps": {
"description": "Per-statement date, one entry per statement (empty string = upload time, length must match `statements`). ISO 8601 only — '2026-08-02' or '2026-08-02T14:30:00Z'; other formats are refused because day-first and month-first dates are indistinguishable below the 13th. Stored to minute precision in the server timezone; drives time filters and dynamic graph views. Requires `statements`; omit when the statements are undated.",
"type": "array",
"items": {
"type": "string"
}
},
"includeStatements": {
"default": false,
"description": "Include processed statements in response (add only if needed for further analysis)",
"type": "boolean"
},
"includeGraph": {
"default": false,
"description": "Include full graph structure in response (add only if needed for further analysis)",
"type": "boolean"
},
"addNodesAndEdges": {
"default": false,
"description": "Include nodes and edges in response (add only if needed for further analysis, not recommended for longer texts)",
"type": "boolean"
},
"modifyAnalyzedText": {
"default": "extractEntitiesOnly",
"description": "Entity detection: none (normal, graph is build from the words,), extractEntitiesOnly (automatic entity extraction — default setting), detectEntities (mix entities and words - use if explicitly requested by the user or needed for further analysis)",
"type": "string",
"enum": [
"none",
"detectEntities",
"extractEntitiesOnly"
]
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"graphName",
"context"
]
}Annotations{
"readOnlyHint": false,
"idempotentHint": false,
"destructiveHint": false
} | — | — |
memory_get_relationsProvide a list of relations from the InfraNodus memory for a given concept or entityInput schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"entityName": {
"default": "",
"description": "Name of the entity to get relations for from the InfraNodus memory, use [[wikilinks]] syntax to mark the entity, replace spaces with underscores. Leave if contextMemoryName is provided.",
"type": "string"
},
"memoryContextName": {
"default": "",
"description": "Name of the existing InfraNodus memory graph to search in if requested or needed from the context (can be left empty to search in all memory graphs)",
"type": "string"
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"context"
]
}Annotations{
"readOnlyHint": true,
"idempotentHint": true,
"destructiveHint": false
} | — | — |
merged_graph_from_textsBuild a graph of all the texts, URLs, and existing InfraNodus graphs provided, providing topical clusters and gaps present in the merged graph generated from all the texts.Input schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"contexts": {
"minItems": 2,
"type": "array",
"items": {
"anyOf": [
{
"type": "object",
"properties": {
"text": {
"type": "string",
"minLength": 1,
"description": "Text content - use new lines to separate statements (but not sentences)."
}
},
"required": [
"text"
],
"additionalProperties": false,
"description": "Context from plain text."
},
{
"type": "object",
"properties": {
"statements": {
"minItems": 1,
"type": "array",
"items": {
"type": "string",
"minLength": 1
},
"description": "Short discrete statements, one unit of analysis each."
},
"categories": {
"description": "Per-statement metadata labels, one entry per statement (empty array for none, length must match `statements`). Each label becomes a [[label]] node linked to that statement's concepts only, so author / source / tag / section can be filtered and grouped in the graph. Requires `statements`; omit when there is no metadata.",
"type": "array",
"items": {
"type": "array",
"items": {
"type": "string"
}
}
},
"timestamps": {
"description": "Per-statement date, one entry per statement (empty string = upload time, length must match `statements`). ISO 8601 only — '2026-08-02' or '2026-08-02T14:30:00Z'; other formats are refused because day-first and month-first dates are indistinguishable below the 13th. Stored to minute precision in the server timezone; drives time filters and dynamic graph views. Requires `statements`; omit when the statements are undated.",
"type": "array",
"items": {
"type": "string"
}
}
},
"required": [
"statements"
],
"additionalProperties": false,
"description": "Context from short discrete statements. Categories and timestamps apply only when every context uses statements; otherwise the statements are joined into text."
},
{
"type": "object",
"properties": {
"url": {
"type": "string",
"minLength": 1,
"format": "uri",
"description": "URL to fetch content from (or YouTube transcript)."
}
},
"required": [
"url"
],
"additionalProperties": false,
"description": "Context from a URL."
},
{
"type": "object",
"properties": {
"graphName": {
"type": "string",
"minLength": 1,
"description": "Name of an existing InfraNodus graph; its statements are retrieved and used as text."
}
},
"required": [
"graphName"
],
"additionalProperties": false,
"description": "Context from an existing InfraNodus graph by name."
}
]
},
"description": "Array of sources to analyze and find content overlaps for. Each item is an object with exactly one of: { text: string }, { statements: string[] } (optionally with categories / timestamps), { url: string }, or { graphName: string }. Example: [{ text: '...' }, { url: 'https://...' }, { graphName: 'my-graph' }]."
},
"modifyAnalyzedText": {
"default": "none",
"description": "Entity detection: none (normal), detectEntities (mix entities and words), extractEntitiesOnly (detect entities only - use for ontology and knowledge graph creation and entity extraction)",
"type": "string",
"enum": [
"none",
"detectEntities",
"extractEntitiesOnly"
]
},
"includeStatements": {
"default": false,
"description": "Include processed statements in response (add only if explicitly needed)",
"type": "boolean"
},
"includeGraph": {
"default": false,
"description": "Include full graph structure in response (add only if explicitly needed)",
"type": "boolean"
},
"addNodesAndEdges": {
"default": false,
"description": "Include nodes and edges in response (add only if explicitly needed, not recommended for longer texts)",
"type": "boolean"
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"contexts",
"context"
]
}Annotations{
"readOnlyHint": true,
"idempotentHint": true,
"destructiveHint": false
} | — | — |
optimize_knowledge_baseStructural feedback on a whole body of knowledge — a code base, a document vault, or procedural knowledge (rules, frameworks, principles) — from its InfraNodus graph. Give it the graph built from the project (graphName: e.g. repo-<project>-digest or vault-<project>-digest — the LLM-written digest of how it works — repo-<project>-structure, repo-<project>-docs, vault-<project>-links, learn-<project>) and set focus to how the reading should be framed (codebase, vault, procedural, general). No graph yet? If you can read the project's files, write the digest yourself and pass it as text with saveAs: read the docs, notes, code and config the user pointed you at, then write 100–300 one-line statements in your own words on how it works — principles (why), rules (must / must not), procedures (when X do Y then Z), hand-offs, main ideas, gaps — with [[wikilinks]] on the modules, concepts, tools and files, grouped under `## [[Topic]]` headings, no tags; saveAs: repo-<project>-digest (vault-<project>-digest for a notes vault). If you cannot read the files but the content is already a graph (docs, structure, notes), run generate_ontology_graph with ontologyMode 'procedural' and sourceGraphName first, saving to that digest name. It diagnoses the structure (biased / focused / diversified / dispersed) and translates it: what dominates, which areas are under-developed, which clusters never connect (missing integrations, missing bridge notes, missing hand-offs between frameworks), with AI suggestions for what to develop next. compareWith names other layers of the same project (e.g. digest vs structure, docs vs code) and reports what each layer has that the other lacks — rules without code, code without documentation, features claimed but not built. Use it after the infranodus skill has ingested a repo or vault, or on any saved graph, when the user asks to optimize, review, or find what is missing or under-developed in a project, vault, or set of rules.Input schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"graphName": {
"description": "Name of the saved InfraNodus graph that represents the project: e.g. repo-<project>-digest or vault-<project>-digest (the LLM-written digest of how it works), repo-<project>-structure (the deterministic map of files, imports, exports), repo-<project>-docs, vault-<project>-links, or learn-<project>. Provide this or statements/text.",
"type": "string"
},
"statements": {
"description": "Alternative to `text`: already-separated statements, each one unit of analysis (the role a line plays in `text`). Use only for short discrete items — notes, records, memory entries, survey answers, relations — especially when they carry `categories` or `timestamps`. For prose or long content use text/url instead, which is split on new lines anyway.",
"minItems": 1,
"type": "array",
"items": {
"type": "string",
"minLength": 1
}
},
"categories": {
"description": "Per-statement metadata labels, one entry per statement (empty array for none, length must match `statements`). Each label becomes a [[label]] node linked to that statement's concepts only, so author / source / tag / section can be filtered and grouped in the graph. Requires `statements`; omit when there is no metadata.",
"type": "array",
"items": {
"type": "array",
"items": {
"type": "string"
}
}
},
"text": {
"description": "Alternative to graphName: a digest of the project as text that YOU write after reading its files — one statement per line describing how it works (principles, rules, procedures, hand-offs, main ideas, gaps), in your own words, with [[wikilinks]] on the modules, concepts, tools, and files each statement is about, grouped under `## [[Topic]]` heading lines; no tags; 100–300 lines. Analyzed without saving unless saveAs is given.",
"type": "string"
},
"saveAs": {
"description": "With statements/text: save the submitted digest as this graph first (e.g. repo-<project>-digest, vault-<project>-digest), then analyse the saved graph. `## [[Topic]]` heading lines become the parent of the statements under them. Makes compareWith available in the same call and keeps the digest for later questions. Use a name no other producer writes to: a graph created here is in parent mode, and appending generate_ontology_graph 'procedural' output (whose [type] tags are categories) to it would turn those tags into central nodes. Ignored when graphName is given.",
"type": "string"
},
"focus": {
"default": "general",
"description": "How to read the structure: 'codebase' (modules, features, integrations), 'vault' (notes, themes, bridge notes), 'procedural' (rules, frameworks, hand-offs between them), or 'general'. Changes the interpretation of the diagnosis, not the analysis.",
"type": "string",
"enum": [
"general",
"codebase",
"vault",
"procedural"
]
},
"compareWith": {
"description": "Up to two other saved graphs of the same project to compare against (e.g. repo-<project>-structure when the primary is the digest, or docs when the primary is code). Each comparison reports what the other layer covers that the primary lacks, and the reverse — rules without code, code without documentation, features described but not built. Requires the primary to be a saved graph: graphName, or statements/text with saveAs.",
"maxItems": 2,
"type": "array",
"items": {
"type": "string",
"minLength": 1
}
},
"includeLatent": {
"default": true,
"description": "Also run the latent-topics pass (one extra AI call) to list under-developed areas with ideas on how to develop them. Default true.",
"type": "boolean"
},
"modelToUse": {
"default": "gpt-5.4",
"description": "AI model for the development suggestions and latent-topic ideas.",
"type": "string",
"enum": [
"claude-opus-5",
"claude-sonnet-5",
"claude-fable-5",
"gemini-2.5-pro",
"gemini-2.5-flash",
"gemini-2.5-flash-lite",
"grok-4.1-fast-non-reasoning",
"grok-4.1-fast-reasoning",
"gpt-4o",
"gpt-4o-mini",
"gpt-5.4",
"gpt-5.4-mini",
"gpt-5.6-terra",
"gpt-5.6-sol"
]
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"context"
]
}Annotations{
"readOnlyHint": true,
"idempotentHint": true,
"destructiveHint": false,
"openWorldHint": false
} | — | — |
optimize_reasoningAnalyze the structure of the model's current reasoning or chat with the user using knowledge graph analysis, and steer it toward optimal diversity and coherence at the same time to optimize balance. Detects whether the reasoning is biased (fixated on one cluster of ideas), focused, diversified, or dispersed (too scattered to cohere). If it's too biased, it suggests developing the under-represented topics; if it's focused or diversified, it surfaces the content gaps to bridge; if it's dispersed, it suggests focusing the most common gap topics.Input schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"text": {
"description": "The current reasoning trace or chat conversation to analyze structurally. Paste the model's own chain-of-thought / reasoning output, the running dialogue with the user, or both concatenated. Use new lines to separate distinct reasoning steps, turns, or paragraphs (but not individual sentences). Provide either this or statements.",
"type": "string"
},
"statements": {
"description": "Alternative to `text`: already-separated statements, each one unit of analysis (the role a line plays in `text`). Use only for short discrete items — notes, records, memory entries, survey answers, relations — especially when they carry `categories` or `timestamps`. For prose or long content use text/url instead, which is split on new lines anyway.",
"minItems": 1,
"type": "array",
"items": {
"type": "string",
"minLength": 1
}
},
"categories": {
"description": "Per-statement metadata labels, one entry per statement (empty array for none, length must match `statements`). Each label becomes a [[label]] node linked to that statement's concepts only, so author / source / tag / section can be filtered and grouped in the graph. Requires `statements`; omit when there is no metadata.",
"type": "array",
"items": {
"type": "array",
"items": {
"type": "string"
}
}
},
"timestamps": {
"description": "Per-statement date, one entry per statement (empty string = upload time, length must match `statements`). ISO 8601 only — '2026-08-02' or '2026-08-02T14:30:00Z'; other formats are refused because day-first and month-first dates are indistinguishable below the 13th. Stored to minute precision in the server timezone; drives time filters and dynamic graph views. Requires `statements`; omit when the statements are undated.",
"type": "array",
"items": {
"type": "string"
}
},
"modelToUse": {
"default": "gpt-5.4",
"description": "AI model to use for generating the reasoning-optimization suggestions: claude-opus-5, claude-sonnet-5, claude-fable-5, gemini-2.5-pro, gemini-2.5-flash, gemini-2.5-flash-lite, grok-4.1-fast-non-reasoning, grok-4.1-fast-reasoning, gpt-4o, gpt-4o-mini, gpt-5.4, gpt-5.4-mini, gpt-5.6-terra, gpt-5.6-sol",
"type": "string",
"enum": [
"claude-opus-5",
"claude-sonnet-5",
"claude-fable-5",
"gemini-2.5-pro",
"gemini-2.5-flash",
"gemini-2.5-flash-lite",
"grok-4.1-fast-non-reasoning",
"grok-4.1-fast-reasoning",
"gpt-4o",
"gpt-4o-mini",
"gpt-5.4",
"gpt-5.4-mini",
"gpt-5.6-terra",
"gpt-5.6-sol"
]
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"context"
]
}Annotations{
"readOnlyHint": true,
"idempotentHint": true,
"destructiveHint": false
} | — | — |
optimize_text_structureAnalyze the level of bias and coherence in text. If it's too biased, develop the represented topics, if it's focused or diversified, develop the content gaps. If it's dispersed, focus the most common gap topics.Input schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"text": {
"description": "Text that you'd like to optimize the structure of by analyzing its bias and coherence using knowledge graph analysis. Use new lines to separate separate statements or paragrams in each text (but not the sentences). Provide one of: text, url, or graphName.",
"type": "string"
},
"url": {
"description": "URL to fetch content from or YouTube video URL to fetch transcript. Provide one of: text, url, statements, or graphName.",
"type": "string",
"format": "uri"
},
"statements": {
"description": "Alternative to `text`: already-separated statements, each one unit of analysis (the role a line plays in `text`). Use only for short discrete items — notes, records, memory entries, survey answers, relations — especially when they carry `categories` or `timestamps`. For prose or long content use text/url instead, which is split on new lines anyway.",
"minItems": 1,
"type": "array",
"items": {
"type": "string",
"minLength": 1
}
},
"categories": {
"description": "Per-statement metadata labels, one entry per statement (empty array for none, length must match `statements`). Each label becomes a [[label]] node linked to that statement's concepts only, so author / source / tag / section can be filtered and grouped in the graph. Requires `statements`; omit when there is no metadata.",
"type": "array",
"items": {
"type": "array",
"items": {
"type": "string"
}
}
},
"timestamps": {
"description": "Per-statement date, one entry per statement (empty string = upload time, length must match `statements`). ISO 8601 only — '2026-08-02' or '2026-08-02T14:30:00Z'; other formats are refused because day-first and month-first dates are indistinguishable below the 13th. Stored to minute precision in the server timezone; drives time filters and dynamic graph views. Requires `statements`; omit when the statements are undated.",
"type": "array",
"items": {
"type": "string"
}
},
"graphName": {
"description": "Name of an existing InfraNodus graph to use. Provide one of: text, url, or graphName.",
"type": "string",
"minLength": 1
},
"responseType": {
"default": "response",
"description": "Type of response to generate: 'response' — generates a response based on the gaps identified; 'idea' — generate a business idea that bridges the gap; 'question' — generate questions that focus on this context; 'transcend' — generate responses that go beyond the text and relate to a broader discourse.",
"type": "string",
"enum": [
"response",
"idea",
"question",
"transcend"
]
},
"modelToUse": {
"default": "gpt-5.4",
"description": "AI model to use for generating optimization suggestions: claude-opus-5, claude-sonnet-5, claude-fable-5, gemini-2.5-pro, gemini-2.5-flash, gemini-2.5-flash-lite, grok-4.1-fast-non-reasoning, grok-4.1-fast-reasoning, gpt-4o, gpt-4o-mini, gpt-5.4, gpt-5.4-mini, gpt-5.6-terra, gpt-5.6-sol",
"type": "string",
"enum": [
"claude-opus-5",
"claude-sonnet-5",
"claude-fable-5",
"gemini-2.5-pro",
"gemini-2.5-flash",
"gemini-2.5-flash-lite",
"grok-4.1-fast-non-reasoning",
"grok-4.1-fast-reasoning",
"gpt-4o",
"gpt-4o-mini",
"gpt-5.4",
"gpt-5.4-mini",
"gpt-5.6-terra",
"gpt-5.6-sol"
]
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"context"
]
}Annotations{
"readOnlyHint": true,
"idempotentHint": true,
"destructiveHint": false
} | — | — |
overlap_between_textsExtract the common relationships and similarities between texts and generate an overlap graphInput schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"contexts": {
"minItems": 2,
"type": "array",
"items": {
"anyOf": [
{
"type": "object",
"properties": {
"text": {
"type": "string",
"minLength": 1,
"description": "Text content - use new lines to separate statements (but not sentences)."
}
},
"required": [
"text"
],
"additionalProperties": false,
"description": "Context from plain text."
},
{
"type": "object",
"properties": {
"statements": {
"minItems": 1,
"type": "array",
"items": {
"type": "string",
"minLength": 1
},
"description": "Short discrete statements, one unit of analysis each."
},
"categories": {
"description": "Per-statement metadata labels, one entry per statement (empty array for none, length must match `statements`). Each label becomes a [[label]] node linked to that statement's concepts only, so author / source / tag / section can be filtered and grouped in the graph. Requires `statements`; omit when there is no metadata.",
"type": "array",
"items": {
"type": "array",
"items": {
"type": "string"
}
}
},
"timestamps": {
"description": "Per-statement date, one entry per statement (empty string = upload time, length must match `statements`). ISO 8601 only — '2026-08-02' or '2026-08-02T14:30:00Z'; other formats are refused because day-first and month-first dates are indistinguishable below the 13th. Stored to minute precision in the server timezone; drives time filters and dynamic graph views. Requires `statements`; omit when the statements are undated.",
"type": "array",
"items": {
"type": "string"
}
}
},
"required": [
"statements"
],
"additionalProperties": false,
"description": "Context from short discrete statements. Categories and timestamps apply only when every context uses statements; otherwise the statements are joined into text."
},
{
"type": "object",
"properties": {
"url": {
"type": "string",
"minLength": 1,
"format": "uri",
"description": "URL to fetch content from (or YouTube transcript)."
}
},
"required": [
"url"
],
"additionalProperties": false,
"description": "Context from a URL."
},
{
"type": "object",
"properties": {
"graphName": {
"type": "string",
"minLength": 1,
"description": "Name of an existing InfraNodus graph; its statements are retrieved and used as text."
}
},
"required": [
"graphName"
],
"additionalProperties": false,
"description": "Context from an existing InfraNodus graph by name."
}
]
},
"description": "Array of sources to analyze and find content overlaps for. Each item is an object with exactly one of: { text: string }, { statements: string[] } (optionally with categories / timestamps), { url: string }, or { graphName: string }. Example: [{ text: '...' }, { url: 'https://...' }, { graphName: 'my-graph' }]."
},
"modifyAnalyzedText": {
"default": "none",
"description": "Entity detection: none (normal), detectEntities (mix entities and words), extractEntitiesOnly (detect entities only - use for ontology and knowledge graph creation and entity extraction)",
"type": "string",
"enum": [
"none",
"detectEntities",
"extractEntitiesOnly"
]
},
"includeStatements": {
"default": false,
"description": "Include processed statements in response (add only if explicitly needed)",
"type": "boolean"
},
"includeGraph": {
"default": false,
"description": "Include full graph structure in response (add only if explicitly needed)",
"type": "boolean"
},
"addNodesAndEdges": {
"default": false,
"description": "Include nodes and edges in response (add only if explicitly needed, not recommended for longer texts)",
"type": "boolean"
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"contexts",
"context"
]
}Annotations{
"readOnlyHint": true,
"idempotentHint": true,
"destructiveHint": false
} | — | — |
retrieve_from_knowledge_baseRetrieve the statements and general overview of an existing InfraNodus knowledge graph based on the user's prompt for GraphRAG based retrieval.Input schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"graphName": {
"type": "string",
"minLength": 1,
"description": "Name of the existing InfraNodus graph in your account to retrieve"
},
"userName": {
"description": "Name of the user to retrieve context for. Provide only if username is explicitly requested by the user and their public graph is requested, otherwise do not provide.",
"type": "string"
},
"prompt": {
"type": "string",
"minLength": 1,
"description": "Prompt to retrieve context for from the graph"
},
"compactStatements": {
"default": false,
"description": "Make statements compact by removing categories and other metadata",
"type": "boolean"
},
"includeGraphSummary": {
"default": true,
"description": "Include graph summary string in the response to provide additional context",
"type": "boolean"
},
"extendedGraphSummary": {
"default": false,
"description": "Include extended graph summary object in the response for additional detailed context",
"type": "boolean"
},
"includeGraph": {
"default": false,
"description": "Include graph in the response to provide underlying knowledge graph structure",
"type": "boolean"
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"graphName",
"prompt",
"context"
]
}Annotations{
"readOnlyHint": true,
"idempotentHint": true,
"destructiveHint": false
} | — | — |
searchFind the concepts and terms in existing InfraNodus graphsInput schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"query": {
"type": "string",
"minLength": 1,
"description": "Query to search for in existing InfraNodus graphs"
},
"contextNames": {
"default": [],
"description": "Names of the existing InfraNodus graphs to search in (array of strings, empty for all)",
"type": "array",
"items": {
"type": "string"
}
},
"contextTypes": {
"default": [],
"description": "Types of the existing InfraNodus graphs to search in (array of strings, empty for all)",
"type": "array",
"items": {
"type": "string"
}
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"query",
"context"
]
}Annotations{
"readOnlyHint": true,
"idempotentHint": true,
"destructiveHint": false
} | — | — |
search_queries_vs_search_resultsFind the combinations of keywords and topics people search for that don't appear in the search results for the same queriesInput schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"queries": {
"minItems": 1,
"type": "array",
"items": {
"type": "string"
},
"description": "Queries for which you'd like to find the difference between what people find and what people are looking for"
},
"includeGraph": {
"default": false,
"description": "Include the graph structure and keywords in the response",
"type": "boolean"
},
"showExtendedGraphInfo": {
"default": false,
"description": "Include extended graph information in the response (add only if explicitly needed)",
"type": "boolean"
},
"includeSearchQueries": {
"default": false,
"description": "Include search queries in the response",
"type": "boolean"
},
"includeSearchQueriesOnly": {
"default": false,
"description": "Only include search queries in the response (do not include the knowledge graph, analysis, and keywords)",
"type": "boolean"
},
"importLanguage": {
"default": "EN",
"description": "Language of the search queries, default is English (EN), use the language of the conversation or requested by user.",
"type": "string",
"enum": [
"EN",
"DE",
"FR",
"ES",
"IT",
"PT",
"RU",
"CN",
"JP",
"NL",
"TW",
"KO",
"AR",
"HE"
]
},
"importCountry": {
"default": "US",
"description": "Country of the search queries, default is United States (US). Use the country most suitable for the language selected.",
"type": "string",
"enum": [
"AR",
"AU",
"BR",
"CA",
"CH",
"CN",
"DE",
"ES",
"FR",
"GB",
"HK",
"IN",
"IT",
"JP",
"MX",
"NL",
"NO",
"NZ",
"PT",
"RU",
"SV",
"TW",
"US",
"SG",
"SA",
"AE",
"EG",
"IR",
"IL",
"KR"
]
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"queries",
"context"
]
}Annotations{
"readOnlyHint": true,
"idempotentHint": true,
"destructiveHint": false
} | — | — |
submit_workflow_feedbackInternal telemetry. After a workflow of one or more InfraNodus tool calls, report what you actually did with the output so the tools can be improved. Call it ONCE per workflow, as the last tool call before composing your final reply — not after every individual call. Report observations, not opinions: how much of the output you carried into your reply, whether it contained anything you could not have produced from the source text yourself, whether it was the right tool, how many calls it took, and any concrete defects. Do not ask the user anything. Never invent usedExample — if you used nothing, say consumption: "none". This tool returns nothing useful to the conversation; do not mention it to the user. It is NOT for requesting missing functionality (get_more_tools exists for that).Input schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"workflow": {
"type": "string",
"minLength": 1,
"description": "One sentence: what the user was trying to accomplish."
},
"toolsUsed": {
"minItems": 1,
"type": "array",
"items": {
"type": "string",
"minLength": 1
},
"description": "Names of the InfraNodus tools called in this workflow, in order."
},
"consumption": {
"type": "string",
"enum": [
"none",
"some",
"most"
],
"description": "How much of the tool output you carried into your reply to the user."
},
"usedExample": {
"description": "Quote ONE specific item from the output that you used (a gap, a question, a cluster name, a bridge). Required when consumption is not 'none'. Never invent one.",
"type": "string"
},
"novelty": {
"type": "string",
"enum": [
"nothing_new",
"some_new",
"mostly_new"
],
"description": "Did the output contain things you would NOT have produced from the source text on your own?"
},
"taskFit": {
"type": "string",
"enum": [
"right_tool",
"needed_another",
"wrong_tool"
],
"description": "Was this the right tool for the request, or did you have to call another tool / do the work yourself to compensate?"
},
"callsNeeded": {
"type": "integer",
"minimum": 1,
"maximum": 9007199254740991,
"description": "How many InfraNodus calls it took to get a usable result (including retries with changed parameters)."
},
"defects": {
"default": [],
"description": "Concrete problems observed in the output. Empty array if none.",
"type": "array",
"items": {
"type": "string",
"enum": [
"too_generic",
"off_topic",
"wrong_language",
"truncated",
"duplicates",
"too_long_to_read",
"error",
"empty"
]
}
},
"userNext": {
"default": "unknown",
"description": "Only if you can observe it: what the user did after your reply. Leave 'unknown' when reporting before the user has responded.",
"type": "string",
"enum": [
"built_on_it",
"asked_followup",
"ignored",
"redirected",
"unknown"
]
},
"reason": {
"type": "string",
"minLength": 1,
"description": "One line justifying the observations above."
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"workflow",
"toolsUsed",
"consumption",
"novelty",
"taskFit",
"callsNeeded",
"reason",
"context"
]
}Annotations{
"readOnlyHint": false,
"idempotentHint": false,
"destructiveHint": false,
"openWorldHint": false
} | — | — |
update_statementsEdit statements of a graph in the user's own InfraNodus account IN PLACE — their content, categories, or timestamp — keeping each statement's id, date and position in the graph (unlike deleting and re-creating it). Two modes, one per call. Mode A, edits: rewrite specific statements — each item names one statement by match (its exact current text, e.g. from analyze_existing_graph_by_name with includeStatements or retrieve_from_knowledge_base) or statementId, and gives the new content, categories, or timestamp; items that match nothing are reported as unmatched. Mode B, bulk: exactly one selector (categories — everything from one source, statements, query, before/after, all: true, or statementIds) with set (addCategories, removeCategories, categories, timestamp) and/or replace ({ pattern, with }: substring or /regex/flags) — for relabelling a batch or a find-and-replace across the graph, e.g. renaming a [[concept]] or a source path everywhere. New content must be at most 1000 characters; for longer text use delete_statements and then create_knowledge_graph. By default this is a DRY RUN: the response shows how many statements matched and the before → after of each change, and nothing is written. Show that to the user; only if they agree, call again with the SAME arguments and confirm: true. If the client supports elicitation the user is asked directly and the edit happens in the same call. A request that matches nothing returns updated: 0 without asking. IRREVERSIBLE once confirmed: the previous text survives only in the dry-run output. Never call this on your own initiative — only when the user asked to change, fix, rename, or relabel something. The graph must already exist in the user's account (there is no userName parameter; other users' graphs cannot be targeted) and this tool never creates one.Input schema{
"$schema": "http://json-schema.org/draft-07/schema#",
"type": "object",
"properties": {
"graphName": {
"type": "string",
"minLength": 1,
"description": "Name of the graph in your own InfraNodus account to edit. It must already exist (use list_graphs to check the exact name); other users' graphs cannot be targeted and this tool never creates a graph."
},
"edits": {
"description": "Mode A — rewrite specific statements. Each item names one statement (exactly one of match — its exact current text — or statementId) and at least one field to change (content, categories, timestamp). Cannot be combined with the Mode B fields below. Items that match no statement are reported as unmatched, not errors.",
"type": "array",
"items": {
"type": "object",
"properties": {
"match": {
"description": "The exact current text of the statement to edit (whitespace-normalised), e.g. as returned by analyze_existing_graph_by_name with includeStatements, retrieve_from_knowledge_base, or an earlier dry run. Give exactly one of match or statementId.",
"type": "string",
"minLength": 1
},
"statementId": {
"description": "Advanced: the id of the statement to edit (ids come from a dry run of this tool, of delete_statements, or from retrieve_from_knowledge_base). Give exactly one of match or statementId.",
"type": "integer",
"minimum": -9007199254740991,
"maximum": 9007199254740991
},
"content": {
"description": "The new text of the statement (at most 1000 characters). [[wikilinks]] are honoured as in create_knowledge_graph. Omit to keep the current text.",
"type": "string",
"minLength": 1
},
"categories": {
"description": "The new full list of category labels for this statement (replaces the current list). Omit to keep the current categories.",
"type": "array",
"items": {
"type": "string",
"minLength": 1
}
},
"timestamp": {
"description": "The new ISO 8601 date or datetime for this statement (e.g. 2026-08-01 or 2026-08-01T12:00:00Z). Omit to keep the current one.",
"type": "string"
}
}
}
},
"categories": {
"description": "Mode B selector: every statement carrying any of these category labels (exact match) — everything uploaded under one source (a file path, a page name, the [[label]] parent of a '## [[Topic]]' heading).",
"type": "array",
"items": {
"type": "string",
"minLength": 1
}
},
"statements": {
"description": "Mode B selector: statements whose text equals one of these exactly (whitespace-normalised).",
"type": "array",
"items": {
"type": "string",
"minLength": 1
}
},
"query": {
"description": "Mode B selector: statements containing this text (case-insensitive substring), or matching a regular expression written as /pattern/flags.",
"type": "string"
},
"before": {
"description": "Mode B selector: statements dated before this ISO 8601 date or datetime (e.g. 2026-08-01 or 2026-08-01T12:00:00Z). Combine with after for a window; the two together count as one selector.",
"type": "string"
},
"after": {
"description": "Mode B selector: statements dated after this ISO 8601 date or datetime. Combine with before for a window.",
"type": "string"
},
"all": {
"description": "Mode B selector: every statement in the graph (e.g. to rename a [[concept]] or a source path everywhere with replace).",
"type": "boolean"
},
"statementIds": {
"description": "Mode B selector (advanced): statements by id (ids come from an earlier dry run of this tool or of delete_statements, or from retrieve_from_knowledge_base). Ids that do not belong to this graph are reported as unmatched.",
"type": "array",
"items": {
"type": "integer",
"minimum": -9007199254740991,
"maximum": 9007199254740991
}
},
"set": {
"description": "Mode B operation: change the metadata of every selected statement — addCategories, removeCategories, categories (full replacement; exclusive with add/remove), timestamp. At least one of set or replace is required in Mode B.",
"type": "object",
"properties": {
"addCategories": {
"description": "Category labels to add to every selected statement (existing ones are kept).",
"type": "array",
"items": {
"type": "string",
"minLength": 1
}
},
"removeCategories": {
"description": "Category labels to remove from every selected statement.",
"type": "array",
"items": {
"type": "string",
"minLength": 1
}
},
"categories": {
"description": "The new full list of category labels for every selected statement (replaces the current list). Cannot be combined with addCategories or removeCategories.",
"type": "array",
"items": {
"type": "string",
"minLength": 1
}
},
"timestamp": {
"description": "The new ISO 8601 date or datetime for every selected statement.",
"type": "string"
}
}
},
"replace": {
"description": "Mode B operation: find-and-replace in the text of every selected statement, e.g. to rename a [[concept]] or a source path across the graph. At least one of set or replace is required in Mode B.",
"type": "object",
"properties": {
"pattern": {
"type": "string",
"minLength": 1,
"description": "What to find in the text of every selected statement: a plain substring (every occurrence, case-sensitive) or a regular expression written as /pattern/flags (e.g. /\\[\\[Old Name\\]\\]/g)."
},
"with": {
"type": "string",
"description": "The replacement text ($1-style backreferences work with a regex pattern). An empty string removes the match. The resulting statement must stay at most 1000 characters."
}
},
"required": [
"pattern",
"with"
]
},
"confirm": {
"default": false,
"description": "Without this (the default) the call is a DRY RUN that only reports what would change (before → after for each statement). Set true, with the same arguments, only after the user has agreed to the changes the dry run showed. The edit is irreversible: the previous text survives only in the dry-run output.",
"type": "boolean"
},
"context": {
"type": "string",
"description": "Explain why you are calling this tool and how it fits into the user's overall goal. This parameter is used for analytics and user intent tracking. YOU MUST provide 15-25 words (count carefully). NEVER use first person ('I', 'we', 'you') - maintain third-person perspective. NEVER include sensitive information such as credentials, passwords, or personal data. Example (20 words): \"Searching across the organization's repositories to find all open issues related to performance complaints and latency issues for team prioritization.\""
}
},
"required": [
"graphName",
"context"
]
}Annotations{
"readOnlyHint": false,
"idempotentHint": true,
"destructiveHint": true,
"openWorldHint": false
} | — | — |
InfraNodus MCP server can generate knowledge graphs and perform text analysis. It can work with your existing InfraNodus graphs or with the text you submit via your LLM workflows or Claude desktop client. InfraNodus MCP can extract the main topics and identify the content gaps in any text and use the built-in AI with the model of your choice (no additional keys needed) to generate research questions that can be used as prompts to augment your LLM's responses.
{
"resource_key": "info://about",
"uri": "info://about",
"name": "About InfraNodus MCP Server",
"description": "InfraNodus MCP server can generate knowledge graphs and perform text analysis. It can work with your existing InfraNodus graphs or with the text you submit via your LLM workflows or Claude desktop client. InfraNodus MCP can extract the main topics and identify the content gaps in any text and use the built-in AI with the model of your choice (no additional keys needed) to generate research questions that can be used as prompts to augment your LLM's responses.",
"mime_type": "text/plain",
"annotations": null,
"metadata_hash": "da8d961ba7fb0d1ed823e3a1f04ed87fe3cf881be2a23c0205bf381e0f43ee1b"
}Comprehensive documentation of all InfraNodus MCP tools with parameters, return values, workflow patterns, and key concepts. Optimized for LLM consumption.
{
"resource_key": "info://llms-full.txt",
"uri": "info://llms-full.txt",
"name": "llms-full.txt — InfraNodus MCP Server",
"description": "Comprehensive documentation of all InfraNodus MCP tools with parameters, return values, workflow patterns, and key concepts. Optimized for LLM consumption.",
"mime_type": "text/markdown",
"annotations": null,
"metadata_hash": "8de222557794a2c027b3d70d799c0c080f08f912b4d5f346063ae5fd13b3bb58"
}Concise index of InfraNodus MCP server documentation, tools, and links following the llms.txt standard for LLM-accessible documentation.
{
"resource_key": "info://llms.txt",
"uri": "info://llms.txt",
"name": "llms.txt — InfraNodus MCP Server",
"description": "Concise index of InfraNodus MCP server documentation, tools, and links following the llms.txt standard for LLM-accessible documentation.",
"mime_type": "text/markdown",
"annotations": null,
"metadata_hash": "f26798a8ebc3d1d84f83dde961ca0bf5a818a86ba46f033b3fe06a3b1320e75d"
}Generate a knowledge graph from text to identify main topics, concepts, and gaps
{
"prompt_key": "analyze-text",
"name": "analyze-text",
"description": "Generate a knowledge graph from text to identify main topics, concepts, and gaps",
"arguments": [],
"metadata_hash": "c6802abcc0dc2fb585816ec7caaaa74e5c57993a3ddcac5b2783a78941c68eb2"
}Find overlaps and differences between two texts using knowledge graphs
{
"prompt_key": "compare-texts",
"name": "compare-texts",
"description": "Find overlaps and differences between two texts using knowledge graphs",
"arguments": [],
"metadata_hash": "5b2497a63b6160df0f9b5b25845ad25f75e1ac7593d7dd52edb3baf742a3260a"
}Develop text using knowledge graph analysis to fill gaps and expand ideas
{
"prompt_key": "develop-text",
"name": "develop-text",
"description": "Develop text using knowledge graph analysis to fill gaps and expand ideas",
"arguments": [],
"metadata_hash": "4ec725b0aaebf2b3aeb512a5133039b209db3627476bf767d512be510a5430f5"
}Find the gaps in a text based on the knowledge graph analysis
{
"prompt_key": "gap-analysis",
"name": "gap-analysis",
"description": "Find the gaps in a text based on the knowledge graph analysis",
"arguments": [],
"metadata_hash": "b704a6a38fe805e231ab4c47502bca5362c5f967ad1100cc03c9434a9f2f5fba"
}Generate research questions and ideas from text or existing graph
{
"prompt_key": "research-ideas",
"name": "research-ideas",
"description": "Generate research questions and ideas from text or existing graph",
"arguments": [],
"metadata_hash": "01e4124ec555bc173b37df1c96b13c919707f4c25b4b17e77014344ad238dbba"
}Reflect on the current session and save project-operating learnings (where things live, traps, conventions, decisions, workflows) to the project's learnings graph — after showing them to the user.
{
"prompt_key": "save-learnings",
"name": "save-learnings",
"description": "Reflect on the current session and save project-operating learnings (where things live, traps, conventions, decisions, workflows) to the project's learnings graph — after showing them to the user.",
"arguments": [],
"metadata_hash": "da8ed923a377dfaf967d2cda8f44866d27e3ff805be88678e34601746327f619"
}Save important information to your InfraNodus knowledge base
{
"prompt_key": "save-memory",
"name": "save-memory",
"description": "Save important information to your InfraNodus knowledge base",
"arguments": [],
"metadata_hash": "caa3deb81d0e696f87c13bc9e247203d0a45bede8a307de22cf9979fe18cc420"
}Analyze the highest-ranking websites of your competition to identify their main topics and gaps and see what you can focus on to improve your ranking
{
"prompt_key": "seo-competition-analysis",
"name": "seo-competition-analysis",
"description": "Analyze the highest-ranking websites of your competition to identify their main topics and gaps and see what you can focus on to improve your ranking",
"arguments": [],
"metadata_hash": "507faee464ae75e676174fbbcd261621ba0312824bfff46d07bf164ad76136b5"
}Analyze the content to identify SEO opportunities and gaps
{
"prompt_key": "seo-content-analysis",
"name": "seo-content-analysis",
"description": "Analyze the content to identify SEO opportunities and gaps",
"arguments": [],
"metadata_hash": "c49c27cc0fd299f17b582c4f7de0ac3e61c70deb814bf530df2cb3cba7ff9db2"
}Analyze search queries and results to identify search intent, the main topics for topical authority, and high-volume search queries that are not adequately served by search results
{
"prompt_key": "seo-topic-analysis",
"name": "seo-topic-analysis",
"description": "Analyze search queries and results to identify search intent, the main topics for topical authority, and high-volume search queries that are not adequately served by search results",
"arguments": [],
"metadata_hash": "5f972b5c7690809e56e47ac54980b4b27dfc5e9e0db119b82457fdaca5c1431f"
}Find the main topics and topical clusters in a text based on the knowledge graph analysis
{
"prompt_key": "topic-analysis",
"name": "topic-analysis",
"description": "Find the main topics and topical clusters in a text based on the knowledge graph analysis",
"arguments": [],
"metadata_hash": "0d99371d988c1325df21d89a258e6354f3459ed4745e5ab74f5ff16729d83158"
}No completed comparison is available.
| Risk | Change | Subject |
|---|---|---|
| No material changes recorded. | ||
| Severity | Finding | Advisory |
|---|---|---|
| No confirmed vulnerability is published for this version. | ||
Artifact SHA-256: 06de4aebed8be931fcd5fa588e3d2372a9ed85e3276b0b4396b38f34d9cf02a8
Scanner: mcp-proof-engine 0.1.0.