{"$schema":"https://wellknown.network/schemas/agent-record-v1.json","schemaVersion":"1","id":"ag_wsg3cmvft3f5","handle":"mcp-local-vision","url":"https://wellknown.network/agents/mcp-local-vision","links":{"self":"https://wellknown.network/agents/mcp-local-vision/record.json","html":"https://wellknown.network/agents/mcp-local-vision","markdown":"https://wellknown.network/agents/mcp-local-vision/record.md","api":"https://wellknown.network/api/v1/agents/mcp-local-vision","status":"https://wellknown.network/api/v1/agents/mcp-local-vision/status","claim":"https://wellknown.network/agents/mcp-local-vision/claim","claimApi":"https://wellknown.network/api/v1/claims","claimDescriptor":"https://wellknown.network/agents/mcp-local-vision/claim.json","badge":"https://wellknown.network/agents/mcp-local-vision/badge.svg","openapi":"https://wellknown.network/openapi.json","history":"https://wellknown.network/api/v1/agents/mcp-local-vision/history","tools":"https://wellknown.network/api/v1/agents/mcp-local-vision/tools"},"ard":{"identifier":"urn:air::server:mcp-local-vision","type":"application/mcp-server-card+json"},"kind":"mcp_server","declared":{"name":"mcp-local-vision","summary":"MCP Server that enables local agentic flows or inference to capture and to 'see/analyze' images in a continuous loop (for example, when using LMStudio for Agentic Flow and a VLM can retrieve but not actually see the real pixels of the picture and the flow breaks).","description":"# 👁️ mcp-local-vision (Local Vision MCP Tool)\n\n> **Give your local agent eyes. Close the autonomous loop.**\n\n**mcp-local-vision** is the missing link for autonomous local AI. It allows standard MCP agents (Your local agent running on LMStudio/Ollama/vLLM, Claude Desktop, Goose, etc.) to **\"see\"** images, screenshots, and visual assets by routing them back through an call to the same inference engine and same model (in case the initial model was a VLM model of course so that it can see and reason about the image thus enhancing your workflows.\nMost current MCP servers like Playwright, File Access and Screenshot tools are missing the ability to allow the Agentic AI mode to actually see and reason about images and just returning responses that the end user at the GUI level can see but the Agentic Model is blind to, thus hindering the agentic flowIt also supports cloud providers like Anthropic, Gemini, OpenAI, Openrouter etc...\n\n## 🚫 The Problem: The \"Blind Agent\" Paradox\n\nYou have a cutting-edge local agent. You ask it to \"Debug my UI.\"\n\n1.  Agent launches the app.\n2.  Agent takes a screenshot to check for errors.\n3.  **Agent stops.** 🛑\n\nWhy? Because standard MCP tools only *save* screenshots to disk. The agent knows the file *exists*, but it cannot *see* the pixels. It is blind to its own actions.\n\n## ✅ The Solution: The Visual Loop\n\n**mcp-local-vision** bridges the file system and your inference engine.\n\n1.  **Agent** takes a screenshot via Playwright/Puppeteer.\n2.  **Agent** calls `analyze_image(path)`.\n3.  **mcp-local-vision** instantly pipes the raw pixels to your already hosted model (no extra host, VRAM/RAM usag, just efficient routing/loop in case the initial model supports Vision, or choose any other endpoint where a VLM is hosted and it works too.\n4.  **Agent** receives a detailed description (from it-self sometimes! for a full descroption on the image which can allow it to proceed further (\"I see a 404 error on the dashboard\").\n5.  **Agent** takes the nex…","publisher":null,"homepage":null,"repository":null,"version":"0.1.0","license":"MIT","protocols":["mcp"],"tags":["mcp"],"pricing":null,"endpoints":[{"url":"pypi:mcp-local-vision","type":"package_pypi","auth":null,"probeable":false}],"skills":null,"tools":null,"extra":null,"attribution":{"kind":"pypi","name":"pypi","license":"pypi","summary":"pypi","version":"pypi","description":"pypi"}},"derived":{"capabilities":[{"slug":"media.image-understanding","name":"Image Understanding","confidence":1,"provenance":"derived"},{"slug":"infra.browser-automation","name":"Browser Automation","confidence":0.859,"provenance":"derived"}],"categories":["infra","media"],"language":"en"},"observed":{"status":"unknown","statusReason":"Distributed as a package to run locally; no network endpoint to check.","lastOkAt":null,"lastProbedAt":null,"statusComputedAt":null,"reliability30d":null,"latestObservations":[],"tools":null,"package":{"name":"mcp-local-vision","registry":"pypi","observedAt":"2026-10-22T09:24:16.913Z","publishedAt":"2025-12-09T21:40:15.530496Z","latestVersion":"0.1.0"},"toolSurface":null,"endpointFacts":[]},"verification":{"claimed":false,"claimedAt":null,"proofs":[]},"provenance":{"sources":[{"source":"pypi","key":"mcp-local-vision","url":"https://pypi.org/project/mcp-local-vision/","firstSeenAt":"2026-09-10T02:23:25.774Z","fetchedAt":"2026-10-22T09:23:00.165Z","normalizedAt":"2026-10-22T09:23:00.165Z"}]},"firstSeenAt":"2026-09-10T02:23:25.774Z","updatedAt":"2026-10-22T09:24:16.913Z"}