Gives text-only LLM coding agents vision by routing images to a multimodal model and returning detailed textual descriptions. Supports local files, URLs, clipboard, base64, raw bytes, and multiple providers like OpenAI, Anthropic, and Gemini
Detailed security scan evidence is not public for this MCP yet. Public identity, registry metadata, and independently observed protocol inventory remain available.
Installation and connection instructions are shown only when supported by retained package, repository, or endpoint evidence.
Version 0.2.7 declares 1 executable entrypoint.
npm install --save-exact mcp-vision-bridge@0.2.7npx -y -p mcp-vision-bridge@0.2.7 mcp-vision-bridge{
"mcpServers": {
"mcp-vision-bridge": {
"command": "npx",
"args": [
"-y",
"-p",
"mcp-vision-bridge@0.2.7",
"mcp-vision-bridge"
]
}
}
}| Canonical slug | llm-vision-mcp-84f8a45c | Deployment | Local Only |
|---|---|---|---|
| Canonical package | npm:mcp-vision-bridge | Repository | KuaaMU/mcp-vision-bridge |
| First published | — | Latest release | — |
| Last security verification | — | Classification confidence | 90% |
| Publication | Draft | Official distribution | Not verified |
| Channel | Identifier | Current version | Versions | Source |
|---|---|---|---|---|
| npm | mcp-vision-bridge | 0.2.7 | 1 | Repository |
| Package | Version | Published / observed | Inventory | Security scan |
|---|---|---|---|---|
| npmmcp-vision-bridge | 0.2.7Current | Sep 5, 2026 | 1 toolsSucceeded · 0 resources · 0 prompts | Evidence restricted |
Independently scan the exact version your agents use, receive alerts when its risk changes, and investigate every finding with retained version evidence.
No public current-version evidence is available yet.
| Tool | Category | Annotations | Risk |
|---|---|---|---|
analyze_imageAnalyze an image using a multimodal model and return a detailed text description.
The vision model sees the image; the calling agent is text-only and cannot.
Sources for `image` (pick one, or pass an array to analyze several at once):
- "path": absolute or relative path to a local image file (PNG/JPEG/WEBP/GIF)
- URL: http(s) URL to an image on the web or a local server
- "data:...": base64 data URI, e.g. data:image/png;base64,<payload>
- "clipboard": read the image currently copied to the system clipboard
- "recent": auto-find the most recently pasted image in THIS session
- "session": auto-find EVERY image pasted in this session (analyze them all in one call)
- "raw": the string itself is the literal raw image bytes
- ["a.png", "b.png", ...]: array of any of the above — all analyzed in one request
Pick `task` for common jobs (describe | ocr | ui | layout | qa) or pass your own `prompt`.
`detail` defaults to "high" for maximum completeness.
Use `save_to` to write a long description to a file and get back only a path + summary.Input schema{
"type": "object",
"properties": {
"image": {
"anyOf": [
{
"type": "string",
"minLength": 1
},
{
"type": "array",
"items": {
"type": "string",
"minLength": 1
}
}
],
"description": "Image source(s): a file path, URL, data: URI, 'clipboard', 'recent', 'session', 'raw', or an array of any of these to analyze multiple images in one call. 'session' analyzes every image pasted in the current session."
},
"prompt": {
"type": "string",
"description": "Free-form question or instruction about the image. Overrides `task`."
},
"task": {
"type": "string",
"enum": [
"describe",
"ocr",
"ui",
"layout",
"qa"
],
"description": "Common analysis task. Ignored when `prompt` is provided."
},
"detail": {
"type": "string",
"enum": [
"low",
"high"
],
"description": "Desired detail level. Defaults to 'high'."
},
"save_to": {
"type": "string",
"description": "Optional file path (.txt/.md) to write the full description to."
}
},
"required": [
"image"
],
"additionalProperties": false,
"$schema": "http://json-schema.org/draft-07/schema#"
} | — | — |
| Endpoint | Transport | Authentication | Health | Observed |
|---|---|---|---|---|
| No verified remote endpoint is linked. | ||||
Install the selected package version with: npm install --save-exact mcp-vision-bridge@0.2.7
llm-vision-mcp Server exposed 1 tool during independent protocol observation, including analyze_image.
The selected current version does not yet have completed public verification. Unknown does not mean clean or vulnerable.
Curated product and capability guides containing this catalog record.