{"$schema":"https://wellknown.network/schemas/agent-record-v1.json","schemaVersion":"1","id":"ag_fbr98vn5tkwh","handle":"docreadi-mcp","url":"https://wellknown.network/agents/docreadi-mcp","links":{"self":"https://wellknown.network/agents/docreadi-mcp/record.json","html":"https://wellknown.network/agents/docreadi-mcp","markdown":"https://wellknown.network/agents/docreadi-mcp/record.md","api":"https://wellknown.network/api/v1/agents/docreadi-mcp","status":"https://wellknown.network/api/v1/agents/docreadi-mcp/status","claim":"https://wellknown.network/agents/docreadi-mcp/claim","claimApi":"https://wellknown.network/api/v1/claims","claimDescriptor":"https://wellknown.network/agents/docreadi-mcp/claim.json","badge":"https://wellknown.network/agents/docreadi-mcp/badge.svg","openapi":"https://wellknown.network/openapi.json","history":"https://wellknown.network/api/v1/agents/docreadi-mcp/history","tools":"https://wellknown.network/api/v1/agents/docreadi-mcp/tools"},"ard":{"identifier":"urn:air::server:docreadi-mcp","type":"application/mcp-server-card+json"},"kind":"mcp_server","declared":{"name":"docreadi-mcp","summary":"MCP server for DocReadi — document data extraction for finance, in your AI client.","description":"# docreadi-mcp\n\nA **local** [MCP](https://modelcontextprotocol.io) server for **DocReadi** —\ndocument data extraction for finance. It runs on your machine, reads documents\noff your disk, and calls the hosted DocReadi API with your API key, so an AI\nclient (Claude Desktop, Claude Code, Cursor, …) can extract documents and query\nyour DocReadi corpus without leaving the chat.\n\n> **Status: v1 published** — on PyPI as\n> [`docreadi-mcp`](https://pypi.org/project/docreadi-mcp/); run with\n> `uvx docreadi-mcp` (see Install / configure below). Plan + history:\n> `../../MCP_SERVER_PLAN.md`.\n\n## How it works\n\nYour MCP client spawns `docreadi-mcp` as a local subprocess and talks to it over\nstdio. The server is a thin courier — all extraction/storage happens on the\nhosted DocReadi API; the server just translates tool calls into HTTPS requests\nand (for ingestion) reads local files. Your API key lives only in your client\nconfig, never in this repo.\n\n## Install / configure\n\nAdd the server to your MCP client config with your DocReadi API key\n(get one at **docreadi.com → Settings → API keys**):\n\n```jsonc\n{\n  \"mcpServers\": {\n    \"docreadi\": {\n      \"command\": \"uvx\",\n      \"args\": [\"docreadi-mcp\"],\n      \"env\": {\n        \"DOCREADI_API_KEY\": \"dr_live_…\"\n        // \"DOCREADI_BASE_URL\": \"https://docreadi.com\"  // override if self-hosting\n      }\n    }\n  }\n}\n```\n\n| Env var | Required | Default |\n|---|---|---|\n| `DOCREADI_API_KEY` | yes | — |\n| `DOCREADI_BASE_URL` | no | `https://docreadi.com` |\n\n## Tools\n\n| Tool | What it does |\n|---|---|\n| `check_connection` | First-run self-test — is DocReadi reachable and is your API key valid? Never errors; returns a diagnostic. |\n| `extract_document` | Extract structured data from a **local** PDF/JPG/PNG — uploads, waits for extraction, returns the fields + line items + totals + confidence. |\n| `get_document` | Fetch a document's status + extracted data by id. |\n| `extract_adhoc` | One-off, non-destructive extraction of caller-defined fields from a doc…","publisher":null,"homepage":"https://docreadi.com/api/docs-guide","repository":"https://github.com/greenmartian138/doc_intelligence","version":"0.1.1","license":"Proprietary","protocols":["mcp"],"tags":["docreadi","documents","extraction","finance","invoice","mcp","ocr"],"pricing":null,"endpoints":[{"url":"pypi:docreadi-mcp","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":"documents.extraction","name":"Data Extraction","confidence":1,"provenance":"declared"},{"slug":"documents.invoice-processing","name":"Invoice Processing","confidence":1,"provenance":"declared"},{"slug":"documents.ocr","name":"OCR","confidence":1,"provenance":"declared"}],"categories":["documents"],"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":"docreadi-mcp","registry":"pypi","observedAt":"2026-10-27T23:25:37.722Z","publishedAt":"2026-06-29T16:56:36.974949Z","latestVersion":"0.1.1"},"toolSurface":null,"endpointFacts":[]},"verification":{"claimed":false,"claimedAt":null,"proofs":[]},"provenance":{"sources":[{"source":"pypi","key":"docreadi-mcp","url":"https://pypi.org/project/docreadi-mcp/","firstSeenAt":"2026-09-09T14:32:00.898Z","fetchedAt":"2026-10-27T23:24:50.754Z","normalizedAt":"2026-10-27T23:24:50.754Z"}]},"firstSeenAt":"2026-09-09T14:32:00.898Z","updatedAt":"2026-10-27T23:25:37.722Z"}