US administrative law is where agencies turn statutes into binding rules — and it spans three official sources that don't talk to each other. This server stitches them into one workflow: the Federal Register (the daily journal of proposed/final rules and notices), the eCFR (the codified Code of Federal Regulations, full text, point-in-time), and Regulations.gov (rulemaking dockets and the public comments filed on them). Every Federal Register document carries the docket ID and CFR-part handles that chain across sources, so a rule can be traced proposal → comments → final → codified text in one sequence.

The agent sees regulatory verbs — search_rules, get_cfr_section, find_comments — not three API clients. The Federal Register and eCFR tools are keyless; the two Regulations.gov tools need a free api.data.gov key and return an actionable error without it.

Tools

Seven tools across the regulatory workflow. The Federal Register and eCFR tools (1–4, 7) are keyless. The two Regulations.gov tools (5–6) need REGULATIONS_GOV_API_KEY; regulations_list_open_comments runs keyless and only adds comment counts when the key is present.

regulations_search_rules

Search the Federal Register — the daily journal of US proposed rules, final rules, notices, and presidential documents (1994–present). Keyless.

• Full-text query across title and body, or browse by filters alone
• Filter by document type (PRORULE, RULE, NOTICE, PRESDOCU), agencies (Federal Register agency slug), and publication date window (published_after / published_before)
• Each result carries the documentNumber, docketIds, regulationIdNumbers (RIN), and cfrReferences that chain into the document, comment, and codified-text tools
• Pagination up to 100 per page, 50 pages; the Federal Register caps navigation at 5,000 records and the match count at 10,000 — when a result set is larger, the tool flags it and steers you to date-windowing

regulations_get_document

Fetch one Federal Register document by its FR number — the stitching tool of this server. Keyless.

• Full metadata: title, type, agencies, abstract, action, effective/comment dates, RINs
• The output renders a cross-source handles block — the docket ID (→ regulations_get_docket / regulations_find_comments), affected CFR parts (→ regulations_get_cfr_section), and Regulations.gov document ID and comment count — each next to the tool name that consumes it
• include_full_text inlines the document body as plain text; default false returns the body URLs only (final rules can run tens of thousands of words)

regulations_browse_cfr

Explore the codified Code of Federal Regulations via eCFR in two modes. Keyless.

• structure mode walks the CFR tree (all 50 titles, or one title's chapters → parts → sections) to discover a cite when the exact citation is unknown
• search mode runs a full-text query across the codified CFR and returns matching sections with their hierarchy path and a snippet
• title and part scope both modes, so a part surfaced by structure can be searched directly instead of filtering a whole title's hits by eye. A part needs its title in either mode — part numbers repeat across the Code — and is matched exactly, so pass it as eCFR writes it (58, not Part 58 or 058)
• A live search hit's hierarchyPath names the part it sits in (Part 51 — Requirements for Preparation, Adoption, and Submittal of Implementation Plans), so the subject matter is readable without a second call; a mirror hit's path stays structural, because the index stores no level names
• Search is served from the synced local mirror (FTS5) only when the mirror's title coverage can answer the request — a title it does not hold, an all-titles query against a scoped mirror, a date, or a cold deploy all go to the live eCFR search API instead
• Every search result reports source (mirror or live) and sourceScope — which corpus answered and what it covers, so an empty result is never mistaken for "no such regulation"
• date searches the section text in effect that day; eCFR indexes 2017-01-03 onward, and an undated search is pinned to eCFR's current index date rather than spanning every historical version
• Both modes assemble a cfrCite that feeds regulations_get_cfr_section, and both hand appendices back the same way: an appendix field with eCFR's verbatim identifier, and a cite that names the appendix rather than the part around it
• The mirror indexes section text only, so it never matches an appendix — its sourceScope says so, since otherwise an appendix that exists and one that does not both read as zero matches

regulations_get_cfr_section

Read the codified text at a CFR location — one section, a whole part, or one appendix — via eCFR. Keyless.