{"$schema":"https://wellknown.network/schemas/agent-record-v1.json","schemaVersion":"1","id":"ag_yp6vauhmds63","handle":"mcp-server-google-vision","url":"https://wellknown.network/agents/mcp-server-google-vision","links":{"self":"https://wellknown.network/agents/mcp-server-google-vision/record.json","html":"https://wellknown.network/agents/mcp-server-google-vision","markdown":"https://wellknown.network/agents/mcp-server-google-vision/record.md","api":"https://wellknown.network/api/v1/agents/mcp-server-google-vision","status":"https://wellknown.network/api/v1/agents/mcp-server-google-vision/status","claim":"https://wellknown.network/agents/mcp-server-google-vision/claim","claimApi":"https://wellknown.network/api/v1/claims","claimDescriptor":"https://wellknown.network/agents/mcp-server-google-vision/claim.json","badge":"https://wellknown.network/agents/mcp-server-google-vision/badge.svg","openapi":"https://wellknown.network/openapi.json","history":"https://wellknown.network/api/v1/agents/mcp-server-google-vision/history","tools":"https://wellknown.network/api/v1/agents/mcp-server-google-vision/tools"},"ard":{"identifier":"urn:air::server:mcp-server-google-vision","type":"application/mcp-server-card+json"},"kind":"mcp_server","declared":{"name":"mcp-server-google-vision","summary":"MCP server providing OCR capabilities to LLMs via Google Cloud Vision API - Read scanned PDFs, handwritten text, and images","description":"# mcp-server-google-vision\n\nServeur MCP [Model Context Protocol](https://modelcontextprotocol.io/introduction) permettant aux LLMs comme Claude de lire des documents scannés, du texte manuscrit et des images avec l'API Google Cloud Vision.\n\n## Description\n\nCe projet implémente un serveur MCP qui donne des capacités de vision avancées aux modèles de langage. Développé par [Kohen Avocats](https://kohenavocats.com), un cabinet d'avocats parisien, cet outil est utilisé quotidiennement pour traiter des documents juridiques complexes : pièces scannées, courriers manuscrits, PDF mal orientés, etc.\n\n### Le problème résolu\n\nLes LLMs comme Claude excellent dans l'analyse de texte, mais ne peuvent pas nativement :\n- Lire des PDF scannés (images sans couche texte)\n- Déchiffrer l'écriture manuscrite\n- Traiter des documents mal orientés ou inversés\n- Extraire du texte de photos de documents\n\nCe serveur MCP comble cette lacune en fournissant une interface standardisée vers Google Cloud Vision, permettant aux LLMs de \"voir\" et lire n'importe quel document.\n\n### Cas d'usage\n\n- **Cabinets d'avocats** : Lecture de pièces scannées, correspondances manuscrites, documents anciens\n- **Alimentation de RAG** : Extraction de texte pour indexation dans des bases vectorielles\n- **Traitement documentaire** : OCR de masse avec gestion automatique des PDFs multi-pages\n- **Accessibilité** : Transcription de documents pour personnes malvoyantes\n\n### Fonctionnalités clés\n\n- **OCR haute précision** : Reconnaissance de texte imprimé et manuscrit\n- **Support multi-pages** : Traitement parallèle des PDFs jusqu'à 2000 pages\n- **Détection d'orientation** : Lecture correcte même si le document est à l'envers\n- **9 features Vision API** : text, document, labels, faces, objects, logos, landmarks, web, safe_search\n- **Retry intelligent** : Backoff exponentiel avec jitter pour une fiabilité maximale\n- **Unicode robuste** : Gestion des noms de fichiers avec accents et caractères spéciaux\n\n## Prérequis\n\n- Pytho…","publisher":null,"homepage":"https://github.com/KohenAvocats/mcp-server-google-vision#readme","repository":"https://github.com/KohenAvocats/mcp-server-google-vision#readme","version":"1.0.0","license":"MIT","protocols":["mcp"],"tags":["anthropic","claude","document-processing","google-vision","handwriting","legal-tech","llm","mcp","model-context-protocol","ocr","pdf"],"pricing":null,"endpoints":[{"url":"pypi:mcp-server-google-vision","type":"package_pypi","auth":null,"probeable":false}],"skills":null,"tools":null,"extra":null,"attribution":{"kind":"pypi","name":"pypi","license":"pypi","repoUrl":"pypi","summary":"pypi","version":"pypi","description":"pypi","homepageUrl":"pypi"}},"derived":{"capabilities":[{"slug":"media.image-understanding","name":"Image Understanding","confidence":1,"provenance":"derived"},{"slug":"documents.ocr","name":"OCR","confidence":1,"provenance":"declared"},{"slug":"documents.contracts","name":"Contracts & Legal","confidence":0.619,"provenance":"derived"}],"categories":["documents","media"],"language":"fr"},"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-server-google-vision","registry":"pypi","observedAt":"2026-10-22T11:30:31.789Z","publishedAt":"2025-12-24T22:26:26.228273Z","latestVersion":"1.0.0"},"toolSurface":null,"endpointFacts":[]},"verification":{"claimed":false,"claimedAt":null,"proofs":[]},"provenance":{"sources":[{"source":"pypi","key":"mcp-server-google-vision","url":"https://pypi.org/project/mcp-server-google-vision/","firstSeenAt":"2026-09-10T03:20:38.188Z","fetchedAt":"2026-10-22T11:28:05.225Z","normalizedAt":"2026-10-22T11:28:05.225Z"}]},"firstSeenAt":"2026-09-10T03:20:38.188Z","updatedAt":"2026-10-22T11:30:31.789Z"}