{
  "markdown": "# Patent & Literature Search MCP\n\n<!-- mcp-name: io.github.patsnap/patent-literature-search-mcp -->\n\n[![Smithery badge](https://smithery.ai/badge/openpatsnap/patsnap-patent-literature-search)](https://smithery.ai/servers/openpatsnap/patsnap-patent-literature-search)\n\nSearch global patents and scientific literature in natural language,\ndirectly inside Claude, Cursor, or any MCP-compatible AI tool. The MCP\nsupports semantic, keyword, BM25, and structured search, then retrieves\nselected records as AI-readable Markdown.\n\n> **Start free:** Register on [Patsnap Open Platform](https://open.patsnap.com)\n> to receive **10,000 free credits**. No credit card is required.\n\n## About Patsnap\n\nPatsnap is a global innovation intelligence platform covering patents,\nscientific literature, chemistry, life sciences, and other R&D data.\n\nTo explore Patsnap data interactively, try\n[Eureka](https://eureka.patsnap.com), Patsnap's AI-native R&D\nassistant. To access data programmatically, use the MCP servers or\nREST API via [Patsnap Open Platform](https://open.patsnap.com).\n\n## What It Does\n\nPatent & Literature Search MCP gives your AI agent direct access to\nPatsnap's patent and academic-literature databases. A single natural-language\nresearch request can cover both corpora and combine the results for AI\nreasoning.\n\n- **Patent search** — keyword, assignee, inventor, IPC class, legal\n  status, date range, and citation filters\n- **Literature search** — semantic and keyword search across\n  peer-reviewed scientific papers\n- **Cross-domain research** — orchestrate patent and paper searches using the\n  same tool and combine the structured results for AI reasoning\n- **Natural language input** — describe your problem or topic in\n  plain English or Chinese; the MCP constructs the search strategy\n- **Precision filtering** — filter by assignee, inventor, legal\n  status, jurisdiction, date range, and citation count simultaneously\n\n## Data Coverage\n\nThis MCP is backed by Patsnap's global patent and scientific-literature\ncoverage.\n\n### Patent coverage\n\n| Data-source group | Representative sources | Coverage |\n|---|---|---|\n| International and regional authorities | WIPO/PCT, EPO, ARIPO, EAPO, OAPI, GCC | Applications, grants, and regional records |\n| Major national authorities | USPTO, CNIPA, JPO, KIPO, DPMA, INPI, UKIPO, CIPO, IP Australia | Applications, grants, designs, and utility models where applicable |\n| Additional jurisdictions | Patent authorities across Europe, Asia-Pacific, the Americas, the Middle East, and Africa | Bibliographic records and authority-supported document types |\n\nPatent data coverage includes:\n\n| Data type | What it covers |\n|---|---|\n| Bibliography | Publication and application identifiers, titles, dates, applicants, assignees, inventors, and classifications |\n| Full text | Abstracts, claims, descriptions, and other document text |\n| PDF | Original patent-document coverage for supported authorities |\n| Legal events | Status changes and patent lifecycle events |\n| File wrapper | Available prosecution and examination records |\n| License | Available patent licensing records |\n| Litigation | Available patent litigation records |\n\nCoverage is refreshed frequently, with patent data updated daily.\n\n### Scientific-literature coverage\n\n| Source type | Representative data sources and publications |\n|---|---|\n| Biomedical and clinical literature | PubMed, The Lancet, BMJ, JAMA, Blood, Cancer Research, Journal of Clinical Oncology, Gastroenterology |\n| Multidisciplinary journals | Nature, Science, Scientific Reports, PLOS ONE, Proceedings of the National Academy of Sciences |\n| Chemistry and materials science | Journal of the American Chemical Society, Angewandte Chemie, Journal of Chemical Physics, Chemical & Engineering News Archive |\n| Physics and engineering | Physical Review Letters, Physical Review B, Journal of Applied Physics, AIP Conference Proceedings, SPIE Proceedings |\n| Preprints and repositories | arXiv, HAL, SSRN, Zenodo, DOAJ |\n| Research datasets and specialist sources | ENCODE Datasets, Worldwide Protein Data Bank, OSTI, RePEc, GBIF, PsycEXTRA |\n\nLiterature records include bibliographic metadata, abstracts, citation\nmetadata, and PDF coverage. The collection spans journal articles,\npreprints, conference proceedings, repositories, datasets, and other\nnon-patent literature.\n\nSee [Patsnap Global Data Coverage](https://open.patsnap.com/data-coverage)\nfor the current authority-level and publication-level coverage matrix.\n\n## What You Can Search\n\n`patsnap_search` searches either patents or scientific literature in one\ncall. Its current input supports:\n\n| Input | What it does |\n|---|---|\n| `semantic_query` | Searches a natural-language technical description or concept |\n| `search_strategy` | Combines `semantic`, `keyword`, and `filter` strategies |\n| `keywords` | Runs precise keyword and BM25 matching |\n| `source` | Selects the `patent` or `paper` corpus |\n| `filters` | Applies provider-supported structured filters |\n| `sort` | Sorts by relevance or a provider-supported field |\n| `offset`, `limit` | Paginates results; up to 100 results per call |\n\nPatent filters may include assignees, inventors, IPC classifications,\njurisdictions, filing or publication dates, legal status, and citation\nconditions. Exact filter names and operators depend on the current hosted\ntool schema.\n\n## What You Can Retrieve\n\n`patsnap_fetch` retrieves up to 100 selected result URLs or patent\npublication numbers as Markdown.\n\n### Patent modules\n\n| Module | Content that may be returned |\n|---|---|\n| `basic` | Bibliographic data, title, abstract, claims, description, and other available record content |\n| `citation` | Patent citations and non-patent literature references |\n| `legal` | Available patent legal data and events |\n| `family` | Available patent-family information |\n| `include_images` | Signed patent-drawing URLs when available |\n\n### Citation data\n\nThe `citation` module can retrieve citation data for selected patents. Where\navailable, this may include:\n\n- patents cited by the selected patent;\n- patents citing the selected patent;\n- applicant, examiner, or search-authority citations;\n- citation categories such as X, Y, and A;\n- cited non-patent literature; and\n- DOI and other bibliographic metadata for cited literature.\n\nCitation direction and type should be determined from the returned record\nlabels. Availability and completeness vary by jurisdiction and patent.\n\n### Literature records\n\nLiterature records support the `basic` module and may include title,\nabstract metadata, authors, affiliations, publication, publication date,\nDOI, and other available bibliographic identifiers.\n\n## Quick Start\n\n### 1. Get your API key\n\nRegister at [Patsnap Open Platform](https://open.patsnap.com) and\ngenerate a free API key from the\n[Patent & Literature Search MCP page](https://open.patsnap.com/marketplace/mcp-servers/patsnap-search).\nNew accounts include 10,000 free credits. No credit card required.\n\n### 2. Add to Claude Code\n\n```bash\nclaude mcp add --transport http search-tool \\\n  \"https://connect.patsnap.com/2b0355/logic-mcp?apikey=YOUR_API_KEY\"\n```\n\nRestart Claude Code and type `/mcp` to confirm it loaded.\n\n### 3. Add to Cursor or Windsurf\n\nAdd the following to your MCP configuration file:\n\n```json\n{\n  \"mcpServers\": {\n    \"patsnap_patent_literature\": {\n      \"url\": \"https://connect.patsnap.com/2b0355/logic-mcp?apikey=YOUR_API_KEY\",\n      \"type\": \"streamableHttp\"\n    }\n  }\n}\n```\n\nReplace `YOUR_API_KEY` with your key from Patsnap Open Platform.\n\n## Run the Containerized MCP Bridge\n\nThe repository includes a local stdio bridge for environments that\ninstall or deploy MCP servers from source. It exposes static tool\ndefinitions without credentials and forwards tool calls to the hosted\nPatsnap Streamable HTTP service when `PATSNAP_API_KEY` is configured.\n\n### Docker\n\n```bash\ndocker build -t patsnap-patent-literature-search .\ndocker run --rm -i \\\n  -e PATSNAP_API_KEY=YOUR_API_KEY \\\n  patsnap-patent-literature-search\n```\n\nMCP client configuration:\n\n```json\n{\n  \"mcpServers\": {\n    \"patsnap_patent_literature\": {\n      \"command\": \"docker\",\n      \"args\": [\n        \"run\",\n        \"--rm\",\n        \"-i\",\n        \"-e\",\n        \"PATSNAP_API_KEY\",\n        \"ghcr.io/patsnap/patent-literature-search-mcp:latest\"\n      ],\n      \"env\": {\n        \"PATSNAP_API_KEY\": \"YOUR_API_KEY\"\n      }\n    }\n  }\n}\n```\n\nThe GitHub Actions workflow publishes version tags such as `v1.1.0`\nto GHCR as both `1.1.0` and `latest`.\n\nAn optional live upstream check can be run manually from GitHub\nActions. Add `PATSNAP_TEST_API_KEY` as a repository Actions secret,\nstart the workflow with **Run workflow**, and enable\n`run_upstream`. Pull requests never receive or use this secret.\n\n## Tools\n\n- `patsnap_search` — Search either patents or scientific literature in a\n  request. Supports natural-language and semantic queries, precise keyword\n  and BM25-based text search, and filters such as assignee, inventor, IPC\n  classification, legal status, jurisdiction, date range, and citations.\n  A single research request may invoke patent and literature searches and\n  combine their results.\n- `patsnap_fetch` — Retrieve a patent or literature record as Markdown using\n  one or more result URLs. Patent records can also be fetched by publication\n  number and may include bibliographic data, claims, descriptions, drawings,\n  citations, legal data, and patent-family data. Literature records include\n  core bibliographic and abstract metadata.\n\n## Development\n\nRequires Node.js 20 or newer.\n\n```bash\nnpm ci\nnpm test\nnpm start\n```\n\nThe test suite verifies:\n\n- static tool discovery without an API key;\n- argument validation and upstream forwarding;\n- structured MCP error handling;\n- a real stdio `initialize` and `tools/list` handshake.\n\nTo verify the container locally:\n\n```bash\ndocker build -t patent-literature-search-mcp:test .\nnpm run test:docker\n```\n\nThe server writes MCP JSON-RPC only to stdout. Application errors are\nreturned as MCP tool results, and credentials are never included in logs.\n\n## Glama Release\n\nAfter claiming the server on Glama:\n\n1. Open the server's Dockerfile Admin page.\n2. Select the repository-root `Dockerfile`.\n3. Add `PATSNAP_API_KEY` as a secret deployment environment variable.\n4. Use `node src/index.js` as the command if Glama does not infer the\n   Dockerfile `CMD`.\n5. Run **Deploy**, verify the MCP build test, and create a release.\n\n## Example Prompts\n\n```\nWhat is Tesla actually working on in battery technology — show me\ntheir recent patent filings, group by technology area, and identify\nwho is leading each.\n```\n\n```\nWhich GLP-1 receptor agonist patents are expiring before 2028,\nand what does recent literature say about next-generation mechanisms?\n```\n\n```\nMy EV battery loses 40% range in cold weather. What technical\nsolutions exist in patents and academic papers?\n```\n\n## Resources\n\n- [Patsnap](https://www.patsnap.com)\n- [Patsnap Open Platform](https://open.patsnap.com)\n- [Patent & Literature Search MCP](https://open.patsnap.com/marketplace/mcp-servers/patsnap-search)\n- [Eureka](https://eureka.patsnap.com)\n- [All Patsnap MCP Servers](https://github.com/patsnap/mcp)\n\n## License\n\nApache-2.0\n\n---\n\nPowered by [Patsnap](https://www.patsnap.com). Innovate with Confidence.\n",
  "bytes": 11285,
  "sha": "a103006c22078e6c7b2d6f99a487b8be68cc0fa98eac27c79fa803ab26d3f509",
  "repo_slug": "patsnap/patent-literature-search-mcp",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_patsnap_patent_literature_sear_eee1c220/readme"
}