io.github.matematicsolutions/us-eli-mcp

List US federal bills of a given congress and bill type. Args: congress: e.g. ``118`` (2023-2024 session). bill_type: one of ``"hr"``, ``"s"``, ``"hres"``, ``"sres"``, ``"hjres"``, ``"sjres"``, ``"hconres"``, ``"sconres"``. limit: max results (default 20, API caps around 250). Returns: ``{"total": int, "items": [...]}`` - each item carries the citation contract.

Fetch full detail (including latest action/status) for one bill. Args: congress: e.g. ``118``. bill_type: e.g. ``"hr"``. number: bill number, e.g. ``1``. Returns: A dict with ``congress``, ``bill_type``, ``number``, ``title``, ``latest_action_text``, ``latest_action_date``, ``lex_uri``, ``human_readable_citation``, ``source_url``.

List enacted-law packages in a GovInfo collection modified since a given date. Args: collection: a GovInfo collection code, e.g. ``"USCODE"`` (United States Code), ``"CFR"`` (Code of Federal Regulations), ``"FR"`` (Federal Register), ``"STATUTE"`` (Statutes at Large). start_date: ISO 8601 UTC timestamp (e.g. ``"2023-01-01T00:00:00Z"``); only packages modified on/after this date are returned. limit: max results (default 20). Returns: ``{"total": int, "items": [...]}`` - each item carries the citation contract.

Fetch metadata and content-format download links for one GovInfo package. Args: package_id: a GovInfo package id, e.g. ``"USCODE-2023-title50"`` or a Federal Register document id - take these from `us_list_code_packages`. Returns: A dict with ``package_id``, ``collection``, ``title``, ``date_issued``, ``congress``, ``last_modified``, ``package_link``, ``download_links`` (format -> URL, e.g. ``"pdf"``, ``"xml"``, ``"txt"``), ``human_readable_citation``, ``source_url``.

Full-text search over Federal Register documents (1994-present, keyless). Args: term: full-text search query, e.g. ``"artificial intelligence"``. doc_type: optional filter - one of ``"RULE"``, ``"PRORULE"`` (proposed rule), ``"NOTICE"``, ``"PRESDOCU"`` (presidential document). presidential_document_type: optional filter for presidential documents - e.g. ``"executive_order"``, ``"proclamation"``, ``"memorandum"``. limit: max results (default 20). Returns: ``{"total": int, "items": [...]}`` - ``total`` is the API's own count (display-capped at 10 000); each item carries the citation contract.

Fetch full metadata for one Federal Register document. Args: document_number: an FR document number, e.g. ``"2026-13726"`` - take it from `us_search_federal_register`. Returns: A dict with ``document_number``, ``title``, ``doc_type``, ``abstract``, ``publication_date``, ``agencies``, ``fr_citation`` (the official citation, e.g. ``"91 FR 41591"``), ``raw_text_url``, ``pdf_url``, ``executive_order_number``, ``human_readable_citation``, ``source_url``.

Full-text search over 8M+ US court opinions (CourtListener, keyless). The headline value is STATE case law - no other tool in this server covers state courts. Federal courts are searchable too, but dedicated CourtListener MCP wrappers exist for heavy federal case-law work. Args: query: full-text query, e.g. ``"negligence standard of care"``. court: optional CourtListener court id to scope the search, e.g. ``"cal"`` (California Supreme Court), ``"ny"`` (NY Court of Appeals), ``"texapp"`` (Texas Court of Appeals), ``"scotus"``. limit: max results (default 20; the API returns 20 per page). Returns: ``{"total": int, "items": [...]}`` - ``total`` is the API's own count; each item carries the citation contract (real reporter citation when one exists, docket number otherwise - never a fabricated one).

Fetch one opinion cluster by its CourtListener ``cluster_id``. Uses the keyless search endpoint with a ``cluster_id:{id}`` field query (the authenticated ``/clusters/{id}/`` endpoint is deliberately not used). Args: cluster_id: e.g. ``8512159`` - take it from `us_search_case_law`. Returns: A dict with ``cluster_id``, ``case_name``, ``court``, ``court_id``, ``date_filed``, ``docket_number``, ``citations`` (reporter citations), ``status``, ``human_readable_citation``, ``source_url``.

Full-text search over the CURRENT Code of Federal Regulations (eCFR, keyless). Unlike the GovInfo ``CFR`` collection (annual-edition snapshots), the eCFR is amended through days ago and every hit carries the exact CFR hierarchy (title/part/section) plus an excerpt. Args: query: full-text query, e.g. ``"unmanned aircraft"``. limit: max results (default 20). Returns: ``{"total": int, "items": [...]}`` - ``total`` is the API's own count (display-capped at 10 000); each item carries the citation contract (``"{title} CFR § {section}"`` + the public ecfr.gov section page).

Declare what this connector covers, how it is sourced, and what it does NOT cover. Call this before telling a user that the law "does not contain" something, and whenever a search comes back empty: the absence may be a gap in this connector rather than in the law. Every gap carries a fallback saying where to look instead. Returns: ``Coverage`` with families, an as-of note, and a non-empty list of known gaps.

Fetch the amendment history (point-in-time versions) of one CFR section. Args: title: CFR title number, e.g. ``15``. part: part identifier, e.g. ``"744"``. section: section identifier, e.g. ``"744.3"`` - take title/part/section from `us_search_cfr_sections`. Returns: ``{"total": int, "items": [...]}`` - each item is one version with ``date``, ``amendment_date``, ``name``, ``substantive``, ``removed``.