> ## Documentation Index
> Fetch the complete documentation index at: https://docs.the40thieves.dev/llms.txt
> Use this file to discover all available pages before exploring further.

# Sources

Generated by `npm run docs` from `src/sources/registry.ts`. Do not edit by hand.

152 sources across 19 clusters (36 hidden pending a key or config not present in this deployment).

## academic (22)

| Name | Kind | Cluster | Freshness | Auth env | Optional env | Verified | Ingest | Description |
| - | - | - | - | - | - | - | - | - |
| apollo | rest | academic | daily | none | - | 2026-09-01 | allowed | Cambridge Apollo — Cambridge University institutional repository. Theses, working papers, preprints, and faculty research. DSpace REST API v7, no auth required. |
| arxiv | rest | academic | daily | none | - | 2026-09-01 | allowed | arXiv — 2M+ open access preprints: physics, math, CS, biology, economics, statistics. Full HTML text for most papers (2018+). |
| core *(hidden)* | rest | academic | daily | CORE\_API\_KEY | - | 2026-09-01 | allowed | CORE: 57M+ open access research papers with full text. Broadest OA academic aggregator across all disciplines. Requires free CORE\_API\_KEY. |
| crossref | rest | academic | daily | none | CONTACT\_EMAIL | 2026-09-03 | allowed | Crossref: authoritative DOI metadata for 160M+ scholarly records, including reference lists and a BibTeX citation. Works keyless; set CONTACT\_EMAIL for the polite pool (higher rate limit). |
| datacite | rest | academic | daily | none | - | 2026-09-03 | allowed | DataCite: DOI metadata for research datasets and software (Zenodo, Figshare, Dryad, OSF, and more all register here) - one source for the whole data-repository DOI tier. No API key required (3,000 req/5min/IP). |
| dblp | rest | academic | daily | none | - | 2026-09-03 | allowed | DBLP: venue-precise computer science bibliography search. No API key required. Keyless, paced at 1 request/second per dblp's own guidance. |
| doab | rest | academic | daily | none | - | 2026-09-01 | allowed | Directory of Open Access Books — 70k+ peer-reviewed OA academic monographs across all disciplines. Metadata + external PDF links. |
| doaj | rest | academic | daily | none | - | 2026-09-01 | allowed | DOAJ — Directory of Open Access Journals. 12M+ peer-reviewed OA articles from 20,000+ journals. No API key required. |
| eric | rest | academic | daily | none | - | 2026-09-01 | allowed | ERIC — 2M+ education research documents from the US Dept of Education. No API key required. |
| europmc | rest | academic | daily | none | - | 2026-09-01 | allowed | Europe PMC — 43M+ biomedical and life science literature. Full text for open access PMC articles. |
| hal | rest | academic | daily | none | - | 2026-09-03 | allowed | HAL (Hyper Articles en Ligne): 4M+ full-text French/EU academic deposits with a Solr search across titles, abstracts, and full text. No API key required. |
| harvardlib *(hidden)* | rest | academic | daily | none | - | 2026-09-01 | allowed | Harvard LibraryCloud: 20M+ records from Harvard Libraries: books, archives, manuscripts, maps, photographs, theses. Public REST API, no key required. Hidden: api.lib.harvard.edu returns 429 from its load balancer for datacenter clients as of 2026-09; revisit. |
| jinaarxiv | mcp | academic | daily | none | JINA\_API\_KEY | 2026-09-02 | allowed | Jina AI's search\_arxiv tool: arXiv preprint search via the same MCP server as jina. Full text is read via the arxiv source when a result carries an arxiv\_id. Works keyless for connecting; set JINA\_API\_KEY for actual search calls. |
| openalex | rest | academic | daily | none | OPENALEX\_API\_KEY, CONTACT\_EMAIL | 2026-09-01 | allowed | OpenAlex: 200M+ scholarly works. Free replacement for Scopus/Web of Science. Covers all disciplines. No API key required (set CONTACT\_EMAIL for the polite pool, or OPENALEX\_API\_KEY for the paid tier). |
| opencitations | rest | academic | daily | none | OPENCITATIONS\_ACCESS\_TOKEN | 2026-09-03 | allowed | OpenCitations Index v2: an open citation graph. search(doi) returns citing works; read(doi) returns the outgoing reference list. No article text - not ingestible. Optional OPENCITATIONS\_ACCESS\_TOKEN for a higher rate limit. |
| ora | rest | academic | daily | none | - | 2026-09-01 | allowed | Oxford ORA: Oxford University Research Archive. Oxford theses, preprints, working papers, and the Oxford Text Archive (classical texts, TEI/XML). No auth required. |
| osf | rest | academic | daily | none | - | 2026-09-01 | allowed | OSF Preprints — PsyArXiv, SocArXiv, EarthArXiv, engrXiv, and more via Open Science Framework. No API key required. |
| plos | rest | academic | daily | none | - | 2026-09-01 | allowed | PLOS — Public Library of Science journals. 100% open access. Biology, medicine, genetics, computational biology. No API key required. |
| pubmed | rest | academic | daily | none | NCBI\_API\_KEY | 2026-09-03 | allowed | PubMed (NCBI E-utilities): MeSH-aware biomedical literature search. read() returns PMC BioC full text when a PMCID exists, otherwise the PubMed abstract via efetch. Works keyless (3 rps); set NCBI\_API\_KEY for 10 rps. |
| semanticscholar | rest | academic | daily | none | SEMANTIC\_SCHOLAR\_API\_KEY | 2026-09-01 | attribution | Semantic Scholar: 200M+ academic papers. Abstracts always available; OA PDF links for open access papers. Supports library\_recommend. Set SEMANTIC\_SCHOLAR\_API\_KEY for a dedicated rate pool. |
| springer *(hidden)* | rest | academic | daily | SPRINGER\_META\_API\_KEY | SPRINGER\_OA\_API\_KEY | 2026-09-01 | allowed | Springer Nature — 16M+ articles via Meta API (all content, abstracts) + OA API (open access full text). Requires SPRINGER\_META\_API\_KEY and SPRINGER\_OA\_API\_KEY (free at dev.springernature.com). |
| zenodo | rest | academic | daily | none | ZENODO\_API\_KEY | 2026-09-01 | allowed | Zenodo — CERN open repository: papers, datasets, software. 2M+ records across all disciplines. |

## ai\_research (3)

| Name | Kind | Cluster | Freshness | Auth env | Optional env | Verified | Ingest | Description |
| - | - | - | - | - | - | - | - | - |
| hfpapers | rest | ai\_research | daily | none | - | 2026-09-01 | allowed | Hugging Face Papers: a community-curated feed of arXiv papers with upvotes and discussion threads. No API key required. Full text is read via the arxiv source (every id is an arXiv id). |
| huggingface | mcp | ai\_research | daily | none | - | 2026-09-02 | allowed | Hugging Face Hub paper search via its MCP hf\_fs tool (an hf://papers search). Full text is read via the arxiv source (every result id is a bare arXiv id). No API key required. |
| paperswithcode | rest | ai\_research | daily | none | - | 2026-09-03 | allowed | Papers with Code: machine learning papers cross-referenced with their code implementations. Queries the community mirror at paperswithcode.co (the original paperswithcode.com API was retired 2025-07-24). No API key required. |

## archives (4)

| Name | Kind | Cluster | Freshness | Auth env | Optional env | Verified | Ingest | Description |
| - | - | - | - | - | - | - | - | - |
| chroniclingamerica | rest | archives | static | none | - | 2026-09-01 | allowed | Chronicling America (LOC): full OCR text of US newspapers 1770-1963, now served via the unified loc.gov search/item API. |
| loc | rest | archives | daily | none | - | 2026-09-01 | allowed | Library of Congress — US history, maps, newspapers, manuscripts. 170M+ items. Metadata and discovery. |
| nara *(hidden)* | rest | archives | daily | NARA\_API\_KEY | - | 2026-09-01 | allowed | US National Archives (NARA): 32M+ descriptions of historical government records: war diaries, presidential papers, military records, federal agencies. Requires free NARA\_API\_KEY. |
| trove *(hidden)* | rest | archives | daily | TROVE\_API\_KEY | - | 2026-09-01 | forbidden | Trove (NLA Australia): 340M+ items from Australian libraries, newspapers, archives. Requires free TROVE\_API\_KEY. Full text is fetched only for newspaper articles that carry a fulltext link; other categories stay metadata-only. Governed by an NLA data agreement: live calls only, no storage, capped full-text reads per session. |

## culture (8)

| Name | Kind | Cluster | Freshness | Auth env | Optional env | Verified | Ingest | Description |
| - | - | - | - | - | - | - | - | - |
| digitalnz | rest | culture | daily | none | DIGITALNZ\_API\_KEY | 2026-09-01 | allowed | DigitalNZ: New Zealand digital heritage including Māori and Pacific content. Keyless (rate-limited); set DIGITALNZ\_API\_KEY for a higher limit. |
| dpla *(hidden)* | rest | culture | daily | DPLA\_API\_KEY | - | 2026-09-01 | allowed | Digital Public Library of America: 17M+ items from US libraries, archives, and museums. Metadata aggregator. Requires free DPLA\_API\_KEY. |
| europeana *(hidden)* | rest | culture | daily | EUROPEANA\_API\_KEY | - | 2026-09-01 | allowed | Europeana — 50M+ cultural heritage items from European museums, libraries, and archives. Requires free EUROPEANA\_API\_KEY. |
| gallica | rest | culture | daily | none | - | 2026-09-01 | allowed | Gallica/BnF — 5M+ digitized documents from the French national library. Full text available for OCR'd items. |
| nbnorway | rest | culture | daily | none | - | 2026-09-01 | allowed | National Library of Norway — digitized books, manuscripts, and newspapers with OCR text in JSON. No API key required. |
| ndl | rest | culture | daily | none | - | 2026-09-01 | allowed | Japan National Diet Library — 350k+ out-of-copyright digitized books. Metadata and discovery; full text via NDL Digital Collections. |
| smithsonian *(hidden)* | rest | culture | daily | SMITHSONIAN\_API\_KEY | - | 2026-09-01 | allowed | Smithsonian Institution — 14M+ records across all museums: American history, air & space, natural history, art, military artifacts. Requires SMITHSONIAN\_API\_KEY (free at api.data.gov). |
| wdl | rest | culture | static | none | - | 2026-09-01 | allowed | World Digital Library (via LOC) — 19k+ rare and unique cultural heritage items from 200 countries, 8000 BCE–present. Maps, manuscripts, photographs, films, sound recordings. |

## developer (17)

| Name | Kind | Cluster | Freshness | Auth env | Optional env | Verified | Ingest | Description |
| - | - | - | - | - | - | - | - | - |
| codewiki | rest | developer | daily | none | - | 2026-09-01 | allowed | Google Code Wiki — AI-generated continuously updated docs for any GitHub repo. Search indexed repos or fetch architecture overviews, API references, and class diagrams by passing "\{owner}/\{repo}". No auth required. |
| context7 *(hidden)* | rest | developer | daily | none | CONTEXT7\_API\_KEY | 2026-09-01 | allowed | Context7: up-to-date library and framework documentation, indexed by library id (e.g. /reactjs/react.dev). Works keyless; set CONTEXT7\_API\_KEY for a higher rate. |
| context7mcp | mcp | developer | daily | none | CONTEXT7\_API\_KEY | 2026-09-02 | allowed | Context7 documentation search via its own MCP server (resolve-library-id then query-docs). Works keyless; set CONTEXT7\_API\_KEY for a higher rate. Additive to the existing context7 REST source. read() cannot target a specific topic (query-docs needs one but read(id) carries no query), so it uses a generic default query; search() again with a specific topic for targeted docs. |
| depsdev | rest | developer | daily | none | - | 2026-09-01 | allowed | deps.dev v3: package version metadata (licenses, dependency links, advisories) across npm, PyPI, cargo, Go, Maven and more. Query as system:name (e.g. npm:lodash); a bare name defaults to npm. |
| devto | rest | developer | realtime | none | - | 2026-09-01 | allowed | DEV Community (dev.to) articles. A single-word query filters by tag; a multi-word query uses the site search endpoint, which returns no results for any query as of 2026-09-01 (an upstream issue, see the task report). No API key required. |
| ecosystems *(hidden)* | rest | developer | daily | CONTACT\_EMAIL | - | - | allowed | Ecosyste.ms Packages: exact package name lookup across every open source package registry it indexes (npm, PyPI, crates.io, RubyGems, and more). Requires CONTACT\_EMAIL (sent as a courteous User-Agent mailto). |
| githubmcp *(hidden)* | mcp | developer | realtime | GITHUB\_TOKEN | - | 2026-09-02 | allowed | GitHub code search via the official Copilot MCP server (search\_code, get\_file\_contents). Requires GITHUB\_TOKEN; hidden without it. |
| githubsearch *(hidden)* | rest | developer | realtime | GITHUB\_TOKEN | - | - | allowed | GitHub code search: full-text search across public repository contents. Requires GITHUB\_TOKEN (the /search/code endpoint rejects unauthenticated requests). |
| hn | rest | developer | realtime | none | - | 2026-09-01 | allowed | Hacker News, via the Algolia HN Search API: stories and comments. No API key required. |
| lobsters | rss | developer | realtime | none | - | 2026-09-01 | allowed | Lobsters: a computing-focused link aggregation and discussion community, as an RSS feed. |
| mdn *(hidden)* | rest | developer | daily | none | - | 2026-09-01 | allowed | MDN Web Docs search: web platform reference and guides. No API key required. |
| mdnmcp | mcp | developer | daily | none | - | 2026-09-02 | allowed | MDN Web Docs's own MCP server: web platform reference and guides, with full document text via get-doc. No API key required. Additive to the existing mdn REST source, which it falls back to when the MCP server is unreachable. |
| peps | rest | developer | daily | none | - | 2026-09-01 | allowed | Python Enhancement Proposals (PEPs): index of every PEP by number and title. Downloaded once per process and filtered client-side; there is no per-query search API. No API key required. |
| readthedocs | rest | developer | daily | none | - | 2026-09-03 | allowed | Read the Docs: server-side search across hosted project documentation. Scope a query to one project with a "project:\<slug>" prefix (e.g. "project:requests authentication") - an unscoped query returns nothing. No API key required for public projects. |
| stackexchange | rest | developer | realtime | none | STACKEXCHANGE\_KEY | 2026-09-01 | attribution | Stack Overflow (via the Stack Exchange API): advanced full-text question search. Works keyless at the shared rate limit; set STACKEXCHANGE\_KEY for a dedicated pool. |
| swiftevolution | rest | developer | daily | none | - | 2026-09-01 | allowed | Swift Evolution proposals: every accepted, rejected, or in-review Swift language proposal with its status. Downloaded once per process and filtered client-side; there is no per-query search API. No API key required. |
| tc39 | rest | developer | daily | none | - | 2026-09-01 | allowed | TC39 (ECMAScript) proposals: every JavaScript language proposal with its current stage. Downloaded once per process and filtered client-side; there is no per-query search API. No API key required. |

## economics (11)

| Name | Kind | Cluster | Freshness | Auth env | Optional env | Verified | Ingest | Description |
| - | - | - | - | - | - | - | - | - |
| bls | rest | economics | daily | none | BLS\_API\_KEY | 2026-09-03 | allowed | BLS (Bureau of Labor Statistics) Public Data API v2: US labor and price statistics (unemployment, CPI, PPI, wages, productivity, JOLTS). No text search exists upstream; search() matches a bundled list of common series. Works keyless (conservative daily cap); set BLS\_API\_KEY for the registered 500/day tier. |
| census | rest | economics | daily | none | - | 2026-09-01 | allowed | US Census Bureau dataset catalog: metadata for every dataset the Census Bureau API exposes (ACS, decennial census, economic surveys, and more). Downloaded once per process and filtered client-side; there is no per-query search API. CENSUS\_API\_KEY is optional and not required for this catalog. |
| comtrade | rest | economics | daily | none | UN\_COMTRADE\_KEY | 2026-09-01 | allowed | UN Comtrade: international merchandise trade statistics by HS (Harmonized System) commodity code. The query must be an HS code (e.g. "0101" or "010121"); any other query returns no results. Uses the keyless public preview endpoint by default. |
| dbnomics | rest | economics | daily | none | - | - | allowed | DBnomics: an aggregator of macroeconomic and statistical datasets from central banks, statistical agencies, and international organizations worldwide. No API key required. |
| eia *(hidden)* | rest | economics | daily | EIA\_API\_KEY | - | - | allowed | US Energy Information Administration (EIA) API v2: energy production, consumption, and pricing series. A dotted query (e.g. "PET.MCRFPUS2.M") looks up a series directly; otherwise search() filters EIA's route catalog by name (v2 has no keyword-search endpoint). Requires free EIA\_API\_KEY. |
| epatri | rest | economics | daily | none | - | 2026-09-01 | allowed | EPA Toxics Release Inventory (TRI): facility search by name, part of EPA's tracking of industrial toxic chemical releases. No API key required. |
| frankfurter | rest | economics | daily | none | - | 2026-09-01 | allowed | Frankfurter: free daily exchange rates from the European Central Bank. A query like "USD" or "USD EUR" looks up live rates; anything else searches currency names. No API key required. |
| fred *(hidden)* | rest | economics | daily | FRED\_API\_KEY | - | - | allowed | FRED: the Federal Reserve Bank of St. Louis's economic data API, hundreds of thousands of US and international time series. Requires free FRED\_API\_KEY. |
| openfda | rest | economics | daily | none | OPENFDA\_API\_KEY | 2026-09-01 | allowed | openFDA drug enforcement reports: FDA drug recalls, searchable by product description. Works keyless; set OPENFDA\_API\_KEY for a higher rate limit. |
| secedgar *(hidden)* | rest | economics | daily | CONTACT\_EMAIL | - | 2026-09-03 | allowed | SEC EDGAR full-text search: US public company filings (10-K, 10-Q, 8-K, and more) since 2001. Requires CONTACT\_EMAIL for the mandatory identifying User-Agent (SEC 403s unidentified requests); 10 rps hard cap. |
| worldbank | rest | economics | static | none | - | 2026-09-03 | allowed | World Bank Indicators API v2: 29.5k+ development indicators (GDP, poverty, health, education, and more) across every country. No server-side search - search() filters the full indicator catalog client-side by name/description match. No API key required. |

## geopolitical (3)

| Name | Kind | Cluster | Freshness | Auth env | Optional env | Verified | Ingest | Description |
| - | - | - | - | - | - | - | - | - |
| hapi *(hidden)* | rest | geopolitical | daily | HDX\_APP\_IDENTIFIER | - | - | allowed | HDX HAPI (Humanitarian Data Exchange, Humanitarian API): conflict-event and coordination data. A country-name query resolves to a conflict-events lookup; other queries search HAPI's location metadata. Requires free HDX\_APP\_IDENTIFIER (self-serve). |
| reliefweb *(hidden)* | rest | geopolitical | daily | RELIEFWEB\_APPNAME | - | - | allowed | ReliefWeb: humanitarian reports and situation updates from UN OCHA and partner organizations. Requires a registered RELIEFWEB\_APPNAME. |
| ucdp *(hidden)* | rest | geopolitical | daily | UCDP\_TOKEN | - | - | allowed | UCDP (Uppsala Conflict Data Program) Georeferenced Event Dataset: individually coded events of organized violence, searchable by country. Requires free UCDP\_TOKEN. |

## government (7)

| Name | Kind | Cluster | Freshness | Auth env | Optional env | Verified | Ingest | Description |
| - | - | - | - | - | - | - | - | - |
| congress *(hidden)* | rest | government | daily | DATA\_GOV\_API\_KEY | - | - | allowed | congress.gov API v3: US federal bill search and full text. There is no /v3/search endpoint and /v3/bill's q= parameter is ignored, so search() filters the most recently updated bills by title token. Shares the api.data.gov key family: reads DATA\_GOV\_API\_KEY first, GOVINFO\_API\_KEY second. |
| datagov | rest | government | daily | none | - | 2026-09-01 | allowed | Data.gov — US government open data catalog. 300k+ datasets from federal, state, local, and tribal agencies. No key required. Metadata and discovery. |
| federalregister | rest | government | daily | none | - | 2026-09-01 | allowed | Federal Register: the daily journal of the US federal government (rules, proposed rules, and notices). No API key required. |
| govinfo *(hidden)* | rest | government | daily | GOVINFO\_API\_KEY | - | 2026-09-01 | allowed | GovInfo: US Congressional Record, Federal Register, US Code, Bills, CFR, and more. GPO official archive. Requires free GOVINFO\_API\_KEY. |
| hansard | rest | government | daily | none | - | 2026-09-03 | allowed | UK Parliament Hansard: full-text search over Commons and Lords debates, with the complete member-by-member record of any one debate. No API key required. |
| regulations *(hidden)* | rest | government | daily | DATA\_GOV\_API\_KEY | - | - | allowed | Regulations.gov API v4: US federal rulemaking dockets and public comment documents. Shares the api.data.gov key family: reads DATA\_GOV\_API\_KEY first, GOVINFO\_API\_KEY second. |
| ukparliament | rest | government | daily | none | - | - | allowed | UK Parliament Bills API: public legislation before the House of Commons and House of Lords. No API key required. |

## law (4)

| Name | Kind | Cluster | Freshness | Auth env | Optional env | Verified | Ingest | Description |
| - | - | - | - | - | - | - | - | - |
| courtlistener *(hidden)* | rest | law | daily | COURTLISTENER\_API\_KEY | - | 2026-09-01 | allowed | CourtListener: US federal and state court opinions. Free Law Project. Requires free COURTLISTENER\_API\_KEY (125 req/day authenticated cap since 2026-05-07). |
| eurlex | rest | law | daily | none | - | 2026-09-03 | attribution | EUR-Lex (via the CELLAR SPARQL endpoint): EU treaties, regulations, directives, and national implementing measures, searchable by title keyword. No API key required. |
| legislation | rest | law | daily | none | - | 2026-09-01 | allowed | legislation.gov.uk — UK Acts of Parliament, Statutory Instruments, and devolved legislation with time-aware full text. No API key required. |
| legislationscot | rest | law | daily | none | - | 2026-09-01 | allowed | Scottish Parliament legislation — Acts of the Scottish Parliament (ASP) and Scottish Statutory Instruments (SSI). No API key required. |

## literature (16)

| Name | Kind | Cluster | Freshness | Auth env | Optional env | Verified | Ingest | Description |
| - | - | - | - | - | - | - | - | - |
| archive | rest | literature | static | none | - | 2026-09-01 | allowed | Internet Archive — 41M+ texts, scanned books, historical documents. |
| ccel | scrape | literature | static | none | - | 2026-09-01 | allowed | Christian Classics Ethereal Library — patristics, Reformation theology, and Christian philosophy. Augustine, Aquinas, Calvin, Luther, Boethius, Origen, Edwards. |
| cervantes | rest | literature | static | none | - | 2026-09-01 | allowed | Cervantes Virtual Library: Spanish and Portuguese literature. Search via the SPARQL linked-data catalog (data.cervantesvirtual.com); full text available for a small curated set of well-known works (Quijote, Lorca, Neruda, Borges, Rulfo, Calderón, San Juan de la Cruz). |
| ctext | rest | literature | static | none | CTEXT\_API\_KEY | 2026-09-01 | allowed | Chinese Text Project: pre-Qin and Han dynasty classical Chinese texts with English translations. Full-text read requires a registered IP or optional CTEXT\_API\_KEY per ctext.org/tools/api; search stays keyless. |
| earlyprint *(hidden)* | rest | literature | static | none | - | 2026-09-01 | allowed | EarlyPrint: 60,000+ Early Modern English texts (EEBO/ECCO/Evans TCP), 1473-1700. Linguistically annotated. BlackLab-indexed. Hidden: the eplab.artsci.wustl.edu BlackLab catalog is returning 404 as of 2026-09; re-enable once it returns. |
| googlebooks *(hidden)* | rest | literature | daily | GOOGLE\_BOOKS\_API\_KEY | - | 2026-09-01 | allowed | Google Books: 40M+ books. Full text for public domain titles; preview snippets for in-copyright works. Requires free GOOGLE\_BOOKS\_API\_KEY (the shared keyless quota is exhausted). |
| gutenberg | rest | literature | static | none | - | 2026-09-01 | allowed | Project Gutenberg — 76k+ public domain books, best for pre-1928 literature. |
| internetclassics | scrape | literature | static | none | - | 2026-09-01 | allowed | Internet Classics Archive (MIT) — 440+ classical works in English translation. |
| marxists | scrape | literature | static | none | - | 2026-09-01 | allowed | Marxists Internet Archive — socialist, anarchist, and critical theory texts. Marx, Engels, Luxemburg, Gramsci, Fanon, Benjamin, and 700+ authors. |
| openiti | rest | literature | static | none | GITHUB\_TOKEN | 2026-09-01 | allowed | OpenITI: 10,000+ Islamicate texts in Arabic and Persian (OpenITI mARkdown). Set GITHUB\_TOKEN for full-corpus code search; without it, search falls back to a small set of per-century data repos (GitHub code search now requires auth even for public repos). |
| openlibrary | rest | literature | daily | none | - | 2026-09-01 | allowed | Open Library — 30M+ records. Metadata and discovery; links to Archive.org for text. |
| perseus | rest | literature | static | none | - | 2026-09-01 | allowed | Perseus Digital Library: classical Greek, Latin, and Arabic texts with translations, served via the Scaife reading environment. |
| projectruneberg | scrape | literature | static | none | - | 2026-09-01 | allowed | Project Runeberg — Scandinavian literature and cultural heritage. Ibsen, Strindberg, Lagerlöf, Andersen, Kierkegaard. |
| sacredtexts | scrape | literature | static | none | - | 2026-09-01 | allowed | Sacred-Texts.com: curated registry of religious/philosophical texts: Quran, Sufi corpus, Vedanta, Buddhism, Taoism, Hermeticism, Christian mysticism. The live site is bot-gated (HTTP 403); read() returns metadata only. |
| standardebooks | scrape | literature | static | none | - | 2026-09-01 | allowed | Standard Ebooks — carefully typeset and proofread public domain ebooks. |
| wikisource | rest | literature | daily | none | - | 2026-09-01 | attribution | Wikisource — multilingual library of free-content source texts in 70+ languages. |

## markets (2)

| Name | Kind | Cluster | Freshness | Auth env | Optional env | Verified | Ingest | Description |
| - | - | - | - | - | - | - | - | - |
| coingecko | rest | markets | realtime | none | COINGECKO\_API\_KEY | 2026-09-01 | allowed | CoinGecko: cryptocurrency search across thousands of coins and tokens, with a price/market-cap/description summary read. Works keyless; set COINGECKO\_API\_KEY for a higher rate limit. |
| twelvedata *(hidden)* | rest | markets | realtime | TWELVEDATA\_API\_KEY | - | - | allowed | Twelve Data: stock, ETF, and forex symbol search across global exchanges, with a daily time series read. Requires free TWELVEDATA\_API\_KEY. |

## news\_global (5)

| Name | Kind | Cluster | Freshness | Auth env | Optional env | Verified | Ingest | Description |
| - | - | - | - | - | - | - | - | - |
| gdelt | rest | news\_global | realtime | none | - | - | allowed | GDELT: the Global Database of Events, Language, and Tone. Full-text search across worldwide news coverage monitored by GDELT. No API key required. |
| googlenews | rss | news\_global | realtime | none | - | 2026-09-01 | allowed | Google News: search-by-query RSS aggregator across global news sources. |
| guardian *(hidden)* | rest | news\_global | realtime | GUARDIAN\_API\_KEY | - | - | timeboxed | The Guardian Open Platform: full-text search across Guardian journalism, with article body text. Requires free GUARDIAN\_API\_KEY. |
| newsdata *(hidden)* | rest | news\_global | realtime | NEWSDATA\_API\_KEY | - | - | allowed | newsdata.io: real-time global news article search. Requires free NEWSDATA\_API\_KEY. |
| wikicurrent | rest | news\_global | daily | none | - | 2026-09-01 | allowed | Wikipedia Portal:Current events: the collaboratively edited daily log of world news, parsed from the portal's rendered HTML. No API key required. |

## news\_regional (15)

| Name | Kind | Cluster | Freshness | Auth env | Optional env | Verified | Ingest | Description |
| - | - | - | - | - | - | - | - | - |
| abc-world | rss | news\_regional | realtime | none | - | 2026-09-01 | allowed | ABC News (Australia): world news desk, as an RSS feed. (region: Australia / Asia-Pacific) |
| aljazeera | rss | news\_regional | realtime | none | - | 2026-09-01 | allowed | Al Jazeera: Qatar-based international news network, as an RSS feed. (region: Middle East) |
| allafrica | rss | news\_regional | realtime | none | - | 2026-09-01 | allowed | AllAfrica: pan-African news aggregator, as an RDF feed. (region: Africa) |
| almonitor | rss | news\_regional | realtime | none | - | 2026-09-01 | allowed | Al-Monitor: independent coverage and analysis of the Middle East, as an RSS feed. (region: Middle East) |
| arabnews | rss | news\_regional | realtime | none | - | 2026-09-01 | allowed | Arab News: Saudi Arabia-based English-language daily, as an RSS feed. (region: Middle East) |
| dailymaverick | rss | news\_regional | realtime | none | - | 2026-09-01 | allowed | Daily Maverick: South African news and investigative journalism, as an RSS feed. (region: Africa) |
| dw | rss | news\_regional | realtime | none | - | 2026-09-01 | allowed | Deutsche Welle: German public international broadcaster, as an RDF feed. (region: Europe) |
| folha-en | rss | news\_regional | realtime | none | - | 2026-09-01 | allowed | Folha de S.Paulo: English-language edition, as an RSS feed. (region: Latin America) |
| france24 | rss | news\_regional | realtime | none | - | 2026-09-01 | allowed | France 24: French public international news channel, as an RSS feed. (region: Europe) |
| nhk | rest | news\_regional | realtime | none | - | 2026-09-01 | allowed | NHK World News: Japanese public broadcaster, English-language world news headlines. |
| nikkeiasia | rss | news\_regional | realtime | none | - | 2026-09-01 | allowed | Nikkei Asia: business and economic news from Asia, as an RSS feed. (region: Asia) |
| restofworld | rss | news\_regional | realtime | none | - | 2026-09-01 | allowed | Rest of World: technology news from the Global South, as an RSS feed. (region: Global South) |
| scmp | rss | news\_regional | realtime | none | - | 2026-09-01 | allowed | South China Morning Post: Hong Kong-based news covering China and Asia, as an RSS feed. (region: East Asia) |
| thediplomat | rss | news\_regional | realtime | none | - | 2026-09-01 | allowed | The Diplomat: current affairs magazine covering the Asia-Pacific, as an RSS feed. (region: Asia-Pacific) |
| thehindu-intl | rss | news\_regional | realtime | none | - | 2026-09-01 | allowed | The Hindu: international news desk of the Indian daily, as an RSS feed. (region: South Asia) |

## real\_estate (2)

| Name | Kind | Cluster | Freshness | Auth env | Optional env | Verified | Ingest | Description |
| - | - | - | - | - | - | - | - | - |
| hud *(hidden)* | rest | real\_estate | daily | HUD\_API\_TOKEN | - | - | allowed | HUD Fair Market Rents (FMR): US rental cost benchmarks by area, used to set Section 8 voucher payment standards. The query must be a 5-digit ZIP code or a 2-letter state code; any other query returns no results. Requires a bearer HUD\_API\_TOKEN. |
| rentcast *(hidden)* | rest | real\_estate | daily | RENTCAST\_API\_KEY | - | - | allowed | RentCast: US rental and sale market statistics by ZIP code. Requires free-tier RENTCAST\_API\_KEY (50 requests/month). |

## science (8)

| Name | Kind | Cluster | Freshness | Auth env | Optional env | Verified | Ingest | Description |
| - | - | - | - | - | - | - | - | - |
| bhl *(hidden)* | rest | science | daily | BHL\_API\_KEY | - | 2026-09-01 | allowed | Biodiversity Heritage Library — natural history, botany, zoology literature. Requires free BHL\_API\_KEY. |
| biorxiv | rest | science | daily | none | - | 2026-09-01 | allowed | bioRxiv: biological sciences preprints. No API key required. bioRxiv has no free-text search API; search() lists the last 7 days of postings (paged, capped) and filters client-side by title/abstract match, so it can miss older matches or ones past the scan cap. Use read(doi) directly when you already have a DOI. |
| medrxiv | rest | science | daily | none | - | 2026-09-03 | allowed | medRxiv: clinical and health sciences preprints. No API key required. medRxiv has no free-text search API; search() lists the last 7 days of postings (paged, capped) and filters client-side by title/abstract match, so it can miss older matches or ones past the scan cap. Use read(doi) directly when you already have a DOI. |
| nasa | rest | science | daily | none | - | 2026-09-01 | allowed | NASA NTRS — space science, aeronautics, and engineering technical reports. No API key required. |
| nasaads *(hidden)* | rest | science | daily | NASA\_ADS\_API\_KEY | - | 2026-09-01 | allowed | NASA ADS — Astrophysics Data System. Premier portal for astronomy, astrophysics, and physics research. Requires NASA\_ADS\_API\_KEY (free at ui.adsabs.harvard.edu). |
| nih | rest | science | daily | none | - | 2026-09-01 | allowed | NIH RePORTER — National Institutes of Health funded research projects and abstracts. No API key required. |
| nsf | rest | science | daily | none | - | 2026-09-01 | allowed | NSF Awards — National Science Foundation research grants and abstracts across all scientific disciplines. |
| osti | rest | science | daily | none | - | 2026-09-01 | allowed | DOE OSTI — Department of Energy research: nuclear, energy, physics, materials science. No API key required. |

## security (14)

| Name | Kind | Cluster | Freshness | Auth env | Optional env | Verified | Ingest | Description |
| - | - | - | - | - | - | - | - | - |
| attack | rest | security | static | none | - | 2026-09-01 | allowed | MITRE ATT\&CK Enterprise: adversary tactics and techniques, downloaded once per process as a single STIX bundle and filtered client-side; there is no per-query search API. No API key required. |
| bleepingcomputer | rss | security | realtime | none | - | 2026-09-01 | allowed | BleepingComputer: computer security and technology news, as an RSS feed. |
| circl | rest | security | daily | none | - | 2026-09-01 | allowed | CIRCL Vulnerability Lookup: full-text search across CVE, GHSA and other vulnerability feeds, run by Luxembourg's CERT. No API key required. |
| cisco-psirt | rss | security | realtime | none | - | 2026-09-01 | allowed | Cisco PSIRT: Cisco security advisories, as an RSS feed. |
| cwe | rest | security | static | none | - | 2026-09-01 | allowed | MITRE CWE: the Common Weakness Enumeration, a single-id weakness lookup (e.g. CWE-79 or 79); there is no keyword search endpoint, so any other query returns \[]. No API key required. |
| epss | rest | security | daily | none | - | 2026-09-01 | allowed | FIRST.org EPSS: the Exploit Prediction Scoring System, a probability (0-1) that a CVE will be exploited in the next 30 days. Query as a single CVE id (e.g. CVE-2021-44228); anything else returns \[]. No API key required. |
| euvd | rest | security | daily | none | - | 2026-09-01 | allowed | EUVD: the European Union Vulnerability Database, ENISA's aggregation of CVE, GHSA and vendor advisories. No API key required. |
| exploitdb | rss | security | realtime | none | - | 2026-09-01 | allowed | Exploit-DB: public archive of exploits and vulnerable software, as an RSS feed. |
| ghsa | rest | security | daily | none | GITHUB\_TOKEN | 2026-09-01 | allowed | GitHub Security Advisories: advisories curated by GitHub across public repositories and package ecosystems. Works keyless (60 req/h); set GITHUB\_TOKEN for a higher rate. |
| kev | rest | security | daily | none | - | 2026-09-01 | allowed | CISA Known Exploited Vulnerabilities catalog: CVEs CISA has confirmed are being actively exploited. Downloaded once per process and filtered client-side; there is no per-query search API. |
| msrc | rss | security | realtime | none | - | 2026-09-01 | allowed | Microsoft Security Response Center: Microsoft product security update guide, as an RSS feed. |
| nvd | rest | security | daily | none | NVD\_API\_KEY | 2026-09-01 | allowed | NIST NVD: the National Vulnerability Database, CVE records enriched with CVSS scores and CPE matches. Works keyless at a slow pace; set NVD\_API\_KEY for a faster one. |
| osv | rest | security | daily | none | - | 2026-09-01 | allowed | OSV.dev: the Open Source Vulnerability database aggregating advisories from GitHub, PyPI, npm, RustSec and more. Query as a direct advisory id (CVE-/GHSA-/PYSEC-/RUSTSEC-/GO-prefixed) for one record, or as ecosystem:name (e.g. npm:lodash) for a package advisory list. No API key required. |
| projectzero | rss | security | realtime | none | - | 2026-09-01 | allowed | Google Project Zero: vulnerability research blog, as an Atom feed. |

## standards (3)

| Name | Kind | Cluster | Freshness | Auth env | Optional env | Verified | Ingest | Description |
| - | - | - | - | - | - | - | - | - |
| ietf | rest | standards | static | none | - | 2026-09-01 | allowed | IETF Datatracker: RFC search by title, full text read from rfc-editor.org. No API key required. |
| nist-csrc | rss | standards | realtime | none | - | 2026-09-01 | allowed | NIST Computer Security Resource Center: draft publications currently open for public comment, as an Atom feed. |
| w3c | rest | standards | daily | none | - | 2026-09-01 | allowed | W3C specifications: the first 1000 (of the full catalog), downloaded once per process and filtered client-side by title, since the upstream API ignores its own q= search parameter. No API key required. |

## video (1)

| Name | Kind | Cluster | Freshness | Auth env | Optional env | Verified | Ingest | Description |
| - | - | - | - | - | - | - | - | - |
| youtube *(hidden)* | rest | video | realtime | none | YOUTUBE\_API\_KEY, SUPADATA\_API\_KEY | 2026-09-01 | allowed | YouTube: video search via the official Data API v3 (requires YOUTUBE\_API\_KEY, \~100 searches/day). Transcripts prefer Supadata (SUPADATA\_API\_KEY, a maintained third-party transcript service) when configured; otherwise fall back to an undocumented endpoint since no official transcript API exists for third-party videos, same risk profile as codewiki (can break without notice, ToS gray area for automated access; low-volume personal/research use only). No ingest: transcripts stay read-only, never indexed. |

## web (7)

| Name | Kind | Cluster | Freshness | Auth env | Optional env | Verified | Ingest | Description |
| - | - | - | - | - | - | - | - | - |
| jina | mcp | web | realtime | none | JINA\_API\_KEY | 2026-09-02 | allowed | Jina AI's live web search and page reader MCP server (search\_web, read\_url). Works keyless for connecting; set JINA\_API\_KEY for actual search/read calls (unauthenticated calls return an error result). cluster web, freshness realtime. |
| jinasearch *(hidden)* | rest | web | realtime | JINA\_API\_KEY | - | - | allowed | Jina AI Search (s.jina.ai): web search returning clean per-result text. Requires JINA\_API\_KEY; hidden without it to respect the 20 RPM anonymous cap shared across every caller. read() fetches and extracts the linked page. |
| searxng *(hidden)* | rest | web | realtime | none | - | - | allowed | SearXNG: a self-hosted, privacy-respecting metasearch engine aggregating results across many upstream search engines. Requires SEARXNG\_URL (a deployment-specific instance URL); hidden without it. read() fetches and extracts the linked page. |
| tavily *(hidden)* | rest | web | realtime | TAVILY\_API\_KEY | - | - | allowed | Tavily: a web search API purpose-built for LLM agents and RAG pipelines, with results pre-summarized for relevance. Requires TAVILY\_API\_KEY; hidden without it. read() fetches and extracts the linked page. |
| webfetch | rest | web | realtime | none | - | - | allowed | Direct web fetch: search() always returns \[]; read(id=\<url>) fetches and extracts the text of any http(s) page via the fetch tier (defuddle, then jina, then crawl4ai, then Browser Run, whichever is configured). Use this to read a URL surfaced by another source, or any URL the caller already has. |
| wikidata | rest | web | daily | none | CONTACT\_EMAIL | 2026-09-03 | allowed | Wikidata: entity search (wbsearchentities) across 100M+ structured knowledge-base items. Entity search only, no full-text read - use library\_read for the entity page URL. Works keyless; set CONTACT\_EMAIL for a compliant contact User-Agent. |
| wikipedia | rest | web | daily | none | CONTACT\_EMAIL | 2026-09-03 | attribution | Wikipedia (English): full-text page search plus full article read (summary + mobile-html extraction). CC BY-SA text: attribution is a license condition, not optional. Works keyless; set CONTACT\_EMAIL for a compliant contact User-Agent. |

## Totals

| Cluster | Sources | Hidden |
| - | - | - |
| academic | 22 | 3 |
| ai\_research | 3 | 0 |
| archives | 4 | 2 |
| culture | 8 | 3 |
| developer | 17 | 5 |
| economics | 11 | 3 |
| geopolitical | 3 | 3 |
| government | 7 | 3 |
| law | 4 | 1 |
| literature | 16 | 2 |
| markets | 2 | 1 |
| news\_global | 5 | 2 |
| news\_regional | 15 | 0 |
| real\_estate | 2 | 2 |
| science | 8 | 2 |
| security | 14 | 0 |
| standards | 3 | 0 |
| video | 1 | 1 |
| web | 7 | 3 |
| **Total** | 152 | 36 |


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.