{
  "markdown": "<p align=\"center\">\n  <a href=\"https://meetdewey.com\">\n    <img src=\"./logo.png\" alt=\"Dewey\" width=\"120\" />\n  </a>\n</p>\n\n# @meetdewey/mcp\n\nMCP server for [Dewey](https://meetdewey.com) — search and research your document collections from Claude, Cursor, and any MCP-compatible agent. See the [full API reference](https://meetdewey.com/docs) for details.\n\n## Installation\n\nAdd to your `claude_desktop_config.json` (or equivalent MCP client config):\n\n```json\n{\n  \"mcpServers\": {\n    \"dewey\": {\n      \"command\": \"/bin/sh\",\n      \"args\": [\"-lc\", \"npx -y @meetdewey/mcp\"],\n      \"env\": {\n        \"DEWEY_API_KEY\": \"dwy_live_...\",\n        \"DEWEY_COLLECTION_ID\": \"...\"\n      }\n    }\n  }\n}\n```\n\n> **Note for Windows users:** replace `\"command\": \"/bin/sh\"` and `\"args\": [\"-lc\", \"npx -y @meetdewey/mcp\"]` with `\"command\": \"npx\"` and `\"args\": [\"-y\", \"@meetdewey/mcp\"]` — the login-shell wrapper is only needed on macOS/Linux to pick up Homebrew and nvm PATH entries that Claude Desktop doesn't inherit.\n\n## Environment variables\n\n| Variable | Required | Description |\n|---|---|---|\n| `DEWEY_API_KEY` | Yes | Your Dewey project API key |\n| `DEWEY_COLLECTION_ID` | No | Default collection ID. When set, tools that accept `collection_id` will fall back to this value. |\n| `DEWEY_API_URL` | No | Override the API base URL (default: `https://api.meetdewey.com/v1`) |\n\n## Available tools\n\n### Search & research\n\n| Tool | Description |\n|---|---|\n| `dewey_list_collections` | List all collections in the project, including description and research instructions when set |\n| `dewey_search` | Hybrid semantic + keyword search over chunk content |\n| `dewey_scan_sections` | Lightweight search over section titles and summaries |\n| `dewey_research` | Agentic research query with configurable depth — returns a grounded, cited answer |\n| `dewey_get_section` | Fetch the full Markdown content of a section by ID |\n\n### Document management\n\n| Tool | Description |\n|---|---|\n| `dewey_list_documents` | List documents in a collection with their processing status |\n| `dewey_get_document_sections` | List all sections in a document (table of contents with heading levels and IDs) |\n| `dewey_get_document_markdown` | Fetch the full converted Markdown content of a document |\n| `dewey_retry_document` | Retry a failed document — clears error state and re-queues processing |\n| `dewey_delete_document` | Permanently delete a document and all its derived data |\n\n### Claims & contradictions\n\n| Tool | Description |\n|---|---|\n| `dewey_list_claims` | List extracted factual claims from a collection or specific document, filterable by importance (1–5) |\n| `dewey_list_contradictions` | List detected contradictions — clusters of conflicting claims with explanations and suggested resolutions |\n| `dewey_detect_contradictions` | Trigger an async contradiction detection run across all claims in a collection |\n| `dewey_get_contradiction_run` | Get the status and stats of the latest contradiction detection run (use to poll after `dewey_detect_contradictions`) |\n| `dewey_resolve_contradiction` | Apply a resolution instruction to a contradiction or dismiss it |\n\n### Deduplication\n\n| Tool | Description |\n|---|---|\n| `dewey_detect_duplicates` | Trigger an async deduplication run — identifies near-duplicate documents by measuring shared content across chunks and marks one per cluster as canonical |\n| `dewey_get_duplicate_run` | Get the status and stats of the latest deduplication run (use to poll after `dewey_detect_duplicates`) |\n| `dewey_list_duplicate_groups` | List near-duplicate groups with canonical + members and coverage percentages |\n| `dewey_promote_duplicate_canonical` | Promote a different member to canonical; old canonical becomes near_duplicate |\n| `dewey_disband_duplicate_group` | Disband a group; former members rejoin retrieval as distinct documents |\n\nNon-canonical documents are excluded from retrieval and contradiction detection. Requires `enableDeduplication: true` on the collection (set via `dewey_update_collection` or the Dewey dashboard).\n\n### Collection settings\n\n| Tool | Description |\n|---|---|\n| `dewey_get_collection_stats` | Get document count, storage, section/chunk/claim counts, and processing status breakdown |\n| `dewey_update_collection` | Update collection name, description, research instructions, visibility, and feature flags |\n| `dewey_recompute_summaries` | Re-run AI section summarization across all documents (e.g. after changing the LLM model) |\n| `dewey_recompute_captions` | Re-run AI captioning for all images and tables across all documents |\n| `dewey_delete_collection` | Permanently delete a collection and all its data |\n\n### Research instructions\n\nCollections can have natural-language instructions that are automatically injected into the research system prompt — for example, noting units, preferred sources, or how to handle missing information. They can be set in the Dewey dashboard, via the API, or directly through `dewey_update_collection`. When `dewey_resolve_contradiction` is used to apply a resolution, the suggested instruction is also appended here automatically.\n\n### Tool usage pattern\n\nA typical read-only research workflow:\n\n1. `dewey_list_collections` — discover available collections and understand their purpose\n2. `dewey_scan_sections` — quickly explore document structure to identify relevant sections\n3. `dewey_get_section` — load full content of a specific section\n4. `dewey_search` — retrieve the most relevant chunks for a focused question\n5. `dewey_research` — run a full agentic research query for questions that need multi-step reasoning\n\nFor deeper analysis and curation:\n\n1. `dewey_get_collection_stats` — assess how much content has been processed\n2. `dewey_list_claims` — surface the most important facts across documents\n3. `dewey_list_contradictions` — identify where documents disagree\n4. `dewey_resolve_contradiction` — apply or dismiss each contradiction, updating collection instructions automatically\n\n## License\n\nMIT\n",
  "bytes": 5987,
  "sha": "0d3dc536bb65399d3fce9889f654910b44a048ab3f90a9ea75bc9e71b3cbc0ad",
  "repo_slug": "meetdewey/mcp",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_com_meetdewey_dewey_5d5171d8/readme"
}