{"$schema":"https://wellknown.network/schemas/agent-record-v1.json","schemaVersion":"1","id":"ag_p8k84rdpj2sy","handle":"videre-mcp","url":"https://wellknown.network/agents/videre-mcp","links":{"self":"https://wellknown.network/agents/videre-mcp/record.json","html":"https://wellknown.network/agents/videre-mcp","markdown":"https://wellknown.network/agents/videre-mcp/record.md","api":"https://wellknown.network/api/v1/agents/videre-mcp","status":"https://wellknown.network/api/v1/agents/videre-mcp/status","claim":"https://wellknown.network/agents/videre-mcp/claim","claimApi":"https://wellknown.network/api/v1/claims","claimDescriptor":"https://wellknown.network/agents/videre-mcp/claim.json","badge":"https://wellknown.network/agents/videre-mcp/badge.svg","openapi":"https://wellknown.network/openapi.json","history":"https://wellknown.network/api/v1/agents/videre-mcp/history","tools":"https://wellknown.network/api/v1/agents/videre-mcp/tools"},"ard":{"identifier":"urn:air::server:videre-mcp","type":"application/mcp-server-card+json"},"kind":"mcp_server","declared":{"name":"videre-mcp","summary":"MCP server that bridges vision models to text-only coding models using Florence-2","description":"# videre-mcp\n\n[![CI](https://github.com/Veedubin/Videre-MCP/actions/workflows/ci.yml/badge.svg)](https://github.com/Veedubin/Videre-MCP/actions/workflows/ci.yml)\n[![Python 3.11+](https://img.shields.io/badge/python-3.11%20%7C%203.12%20%7C%203.13-blue.svg)](https://www.python.org/downloads/)\n[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)\n[![Code style: ruff](https://img.shields.io/badge/code%20style-ruff-261230.svg)](https://docs.astral.sh/ruff/)\n\nMCP server that bridges vision models to text-only coding models using Florence-2.\n\nNon-vision LLMs can't see images — videre-mcp fixes that. It loads a Florence-2 vision model locally and exposes six MCP tools that convert images (including SVGs) and screenshots into structured text descriptions that any text-based model can consume.\n\n```\nScreenshot tool → videre-mcp (Florence-2) → Text description → Coding model\n```\n\n## Installation\n\n```bash\npip install videre-mcp\n```\n\nOr with uv:\n\n```bash\nuv pip install videre-mcp\n```\n\nRequires Python 3.11+ and ~300MB disk space for the Florence-2-base model weights (downloaded automatically on first use).\n\n## Usage\n\nAdd to your OpenCode configuration:\n\n```json\n{\n  \"mcpServers\": {\n    \"videre-mcp\": {\n      \"command\": \"videre-mcp\"\n    }\n  }\n}\n```\n\nOr run directly:\n\n```bash\nvidere-mcp\n# or\npython -m videre_mcp\n```\n\n## Model Modes\n\nAll Florence-2 tools support a `model_mode` parameter to balance speed and quality:\n\n- `\"fast\"` (default) — Uses **Florence-2-base**. Fast, lightweight, runs on CPU/GPU.\n- `\"deep\"` — Uses **MiniCPM-V 4.6**. Significantly higher quality for complex visual reasoning.\n  - **Requires:** `pip install videre-mcp[deep]`\n  - **Hardware:** ~8GB VRAM recommended.\n\n## Tools\n\n### `describe_image`\n\nGenerate a natural language description of an image.\n\n**Parameters:**\n- `image_path` (str) — Path to the image file (supports PNG, JPEG, SVG)\n- `detail_level` (str, optional) — `\"normal\"` (default) for brief caption, `\"high\"` for detailed descripti…","publisher":{"name":"videre-mcp contributors","url":null},"homepage":null,"repository":null,"version":"0.3.2","license":null,"protocols":["mcp"],"tags":["docling","florence-2","image-description","mcp","minicpm-v","ocr","paddleocr","screenshot","vision"],"pricing":null,"endpoints":[{"url":"pypi:videre-mcp","type":"package_pypi","auth":null,"probeable":false}],"skills":null,"tools":null,"extra":null,"attribution":{"kind":"pypi","name":"pypi","summary":"pypi","version":"pypi","description":"pypi","publisherName":"pypi"}},"derived":{"capabilities":[{"slug":"media.image-understanding","name":"Image Understanding","confidence":1,"provenance":"declared"},{"slug":"documents.ocr","name":"OCR","confidence":1,"provenance":"declared"}],"categories":["documents","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":"videre-mcp","registry":"pypi","observedAt":"2026-10-22T23:26:14.393Z","publishedAt":"2026-08-24T07:39:52.007244Z","latestVersion":"0.3.2"},"toolSurface":null,"endpointFacts":[]},"verification":{"claimed":false,"claimedAt":null,"proofs":[]},"provenance":{"sources":[{"source":"pypi","key":"videre-mcp","url":"https://pypi.org/project/videre-mcp/","firstSeenAt":"2026-09-10T15:23:36.323Z","fetchedAt":"2026-10-22T23:24:37.235Z","normalizedAt":"2026-10-22T23:24:37.235Z"}]},"firstSeenAt":"2026-09-10T15:23:36.323Z","updatedAt":"2026-10-22T23:26:14.393Z"}