Skip to content

Federal Register MCP server

com.1102tools/federal-register-mcp1102tools.comstreamable-httpv1.0.21Scanned

Federal Register proposed and final rules, notices, executive orders, and FAR cases. 8 tools.

8 tools: 1 change things and 7 only read.

Tools: 0 delete, 0 irreversible, 1 write, 7 read.
  • Reads private data · 6
  • Reads text others write · 8
  • Can send data out · 1
  • 2 findings

Tools listed without sign-in. Remote posture

With Anlyon

1 tool here changes things. Govern it.

  • Anlyon lists the tools and adopts none.
  • Each tool you adopt gets approvals and receipts.
  • Your agent never holds the key.
Govern this server →

A tool your agent calls directly is not routed through Anlyon and is not governed by it. Anlyon Radar only lists and labels, and these labels are the same either way.

What the Federal Register MCP server can do

The Federal Register MCP server is published by 1102tools.com, a domain the MCP Registry verified. It has 8 tools. It can create or change data (1 tool: open_comment_periods), but cannot delete data. 7 tools only read.

6 tools read private data (search_documents, get_document, get_documents_batch and 3 more); 8 tools bring in text other people can write (search_documents, get_document, get_documents_batch and 5 more); 1 tool can send data out of your system (open_comment_periods). On its own it has all three parts of a prompt-injection exfiltration path: private data, text others write, and a way out.

The fixed checks matched 2 findings (long-description and annotation-conflict). Each one comes with the text it matched.

Its remote endpoint lists its tools without sign-in, and its tools changed once since Anlyon Radar first saw them.

Last scanned Nov 9, 2026, version 1.0.21.

Tools

Grouped by what each tool does, most consequential first. Open one for its reasons and full description; descriptions are the publisher's own words.

8 tools

Changes things 1Claude Code: ask
  • open_comment_periodsFind source-reported open comment periods, soonest indexed deadline first.Reads text others writeCan send data out1 finding

    Why it is in this group

    • The name "open_comment_periods" says "open": a word for a change.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    • outboundThe name "open_comment_periods" says "open" and "comment": it writes something other people read.

    Findings

    • Warn

      The server marks "open_comment_periods" read-only (readOnlyHint: true), and its own name, description or input give the class "write".

      Matched text

      annotations: {"readOnlyHint":true,"destructiveHint":false,"openWorldHint":true}; The name "open_comment_periods" says "open": a word for a change.

      What this check looks for
    Description, as the server wrote it
    Find source-reported open comment periods, soonest indexed deadline first. Covers proposed rules, notices, AND final/interim rules that accept comments (dozens of RULE-type documents have open periods at any time). Queries for comment close date >= today, scans up to 500 matching documents oldest-published first (where the soonest deadlines live), sorts by close date, and returns the first `limit`. Source scope: total_open is the API's matching metadata count; scanned is how many this call examined. When total_open exceeds scanned, narrow with agencies/term for complete indexed coverage. A very recently published document with an unusually short comment window can fall outside the scan in that oversubscribed case. complete means the matching indexed records were scanned, not that every currently open opportunity or controlling deadline was verified. The API's comments_close_on metadata can disagree with DATES after an extension or reopening. Even a zero result needs docket/subject search_documents and get_document DATES/linked publication checks before ruling out comments. Default: searches all agencies. Pass agency slugs to narrow scope, e.g. ['defense-acquisition-regulations-system'] for DFARS or ['general-services-administration'] for GSA and GSAR. Parameters: - far_council: True for open FAR proposed rules and FAR information collections (documents filed jointly by DoD, GSA and NASA; there is no single FAR agency slug and the OFPP tag only appears from about mid-2025). Cannot be combined with agencies. - limit: max documents returned after sorting (default 50, max 100). Unfiltered dumps across all agencies can approach 200KB; several broad agencies together (e.g. DoD + GSA) can pass 100KB, so lower limit for those.

    Annotations the server declares

    • destructiveHint: false
    • openWorldHint: true
    • readOnlyHint: true
    • title: Open Comment Periods
Only reads 7Show 7 read-only tools
  • far_case_historyGet all Federal Register documents for a FAR/DFARS case or FAC.Reads private dataReads text others writeNo findings

    Why it is in this group

    • The description's first verb is "get": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "get" and "document": it reads documents.
    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Get all Federal Register documents for a FAR/DFARS case or FAC. Pass a docket ID like 'FAR Case 2023-008' (or a FAC number like 'FAC 2025-06'). Runs BOTH a docket-id search and a quoted full-text search, then merges the two result sets (deduped by document number) in chronological order to show the full rulemaking progression: ANPRM -> proposed rule -> final rule -> corrections. The dual query matters: FAC introduction and companion documents often carry only internal docket numbers, so a docket-only search returns a partial set. Docket matching is token-based, not substring: 'FAR Case 2023' matches all 2023 cases (token '2023' matches '2023-008'), but a partial token like 'FAR Case 20' matches nothing. Be specific to avoid false positives. Each document carries matched_by: 'docket' or 'docket+text' when its docket ids name the case, 'text' when it only mentions the case in its text (regulatory agendas, related or companion cases, FAC introductions, public meetings). Read 'text' matches with care. Each underlying search returns at most 100 documents. truncated=True flags that one of the searches hit that cap (docket_matches and term_matches carry the API's full counts); narrow the docket_id if so. Minimum docket_id length is 3 characters to prevent token matches that return unrelated documents (e.g. 'x' matched 65 random dockets in 0.1.x).

    Annotations the server declares

    • destructiveHint: false
    • openWorldHint: true
    • readOnlyHint: true
    • title: FAR Case History
  • get_documentGet full details for a single Federal Register document by number.Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "get_document" says "get": a word for reading.

    What it can reach

    • private-dataThe name "get_document" says "get" and "document": it reads documents.
    • private-dataThe description's first sentence says "get" and "document": it reads documents.
    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Get full details for a single Federal Register document by number. Returns all available fields including full text URLs, docket info, RIN details, page views, topics, corrections, and CFR references. Document numbers look like '2026-03065' (modern), 'E9-12940' or 'X94-70302' (legacy pre-2011 archive), or 'C1-2026-01234' (corrections).

    Annotations the server declares

    • destructiveHint: false
    • openWorldHint: true
    • readOnlyHint: true
    • title: Get Document
  • get_documents_batchFetch multiple documents in one call (up to ~20).Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "get_documents_batch" says "get": a word for reading.

    What it can reach

    • private-dataThe name "get_documents_batch" says "get" and "document": it reads documents.
    • private-dataThe description's first sentence says "fetch" and "document": it reads documents.
    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Fetch multiple documents in one call (up to ~20). Pass a list of document numbers to fetch up to about 20 in one call. Always returns {count, results, [errors]}; errors.not_found lists any requested numbers the API could not locate. Results come back in the order requested. Image metadata (images, images_metadata) is left out to keep batches small; get_document returns it.

    Annotations the server declares

    • destructiveHint: false
    • openWorldHint: true
    • readOnlyHint: true
    • title: Get Documents Batch
  • get_facet_countsGet document counts grouped by type, agency, topic, or time period.Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "get_facet_counts" says "get": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "get" and "document": it reads documents.
    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Get document counts grouped by type, agency, topic, or time period. Accepts the same filter conditions as search_documents. Returns aggregated counts without individual document results. Facets: type, agency, topic (categorical) plus daily, weekly, monthly, quarterly, yearly (publication-date buckets, useful for trend lines). Useful for understanding the volume of rulemaking by agency, type, or over time within a date range before drilling into specific documents. presidential_document_type (executive_order, proclamation, memorandum, determination, notice, presidential_order, other) narrows to one kind of presidential document, e.g. executive orders per month. At least one filter (agencies, doc_types, term, pub_date_gte/lte, or cfr_title) is required. An unfiltered facet query returns the entire all-time aggregate. FAR Council counts: facets cannot express "filed by DoD AND GSA AND NASA", and no single agency slug covers FAR documents (OFPP only from about mid-2025; 'federal-acquisition-regulation-system' holds 2 documents ever). Count FAR rules with search_documents(far_council=True, doc_types=[...], pub_date_gte/lte=...) and read its count.

    Annotations the server declares

    • destructiveHint: false
    • openWorldHint: true
    • readOnlyHint: true
    • title: Get Facet Counts
  • get_public_inspectionGet current public inspection documents (pre-publication).Reads private dataReads text others writeNo findings

    Why it is in this group

    • The name "get_public_inspection" says "get": a word for reading.

    What it can reach

    • private-dataThe description's first sentence says "get" and "document": it reads documents.
    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    Get current public inspection documents (pre-publication). Public inspection documents are FR documents filed for publication but not yet published. Updated business days only. The current-PI endpoint does NOT support server-side filtering. This tool fetches all current PI documents and filters client-side by agency and/or keyword in the title. Useful for getting early notice of upcoming regulatory actions. Parameters: - agency_filter: case-insensitive substring match against agency slugs, names, and short names. Sub-agencies are included: PI documents list only the FILING sub-agency, so the filter is also matched against the Federal Register agency list and every agency below a matching one counts ('defense-department' or 'DOD' also returns Army, Navy, Air Force, Corps of Engineers and Defense Logistics Agency filings). filters_applied.includes_sub_agencies says whether that worked. - keyword_filter: substring match against document titles - limit: max documents returned after filtering (default 50, max 500). Unfiltered dumps can exceed 170KB; narrow with filters or raise the cap.

    Annotations the server declares

    • destructiveHint: false
    • openWorldHint: true
    • readOnlyHint: true
    • title: Get Public Inspection
  • list_agenciesList agencies with their IDs, names, slugs, and parent agencies.Reads text others writeNo findings

    Why it is in this group

    • The name "list_agencies" says "list": a word for reading.

    What it can reach

    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.
    Description, as the server wrote it
    List agencies with their IDs, names, slugs, and parent agencies. Use the 'slug' values with search_documents() and other tools. Common procurement slugs: - defense-department (DoD) - general-services-administration (GSA) - national-aeronautics-and-space-administration (NASA) - defense-acquisition-regulations-system (DARS/DFARS) - federal-procurement-policy-office (OFPP and the CAS Board) - small-business-administration (SBA) - veterans-affairs-department (VA) The FAR has no single slug. FAR rules and notices are filed jointly under DoD, GSA and NASA; OFPP was added only from about mid-2025, so the OFPP slug misses earlier FAR rules. The agency named 'Federal Acquisition Regulation System' (federal-acquisition-regulation-system) holds 2 documents ever; it carries a note in these results. For FAR documents use search_documents(far_council=True) or open_comment_periods(far_council=True). Parameters: - query: optional case-insensitive substring match against name, short_name, and slug. Recommended: narrow results before pulling full detail. - include_detail: if False (default), returns only id/name/short_name/slug/parent_id. If True, returns all fields (description, urls, etc.). The full dump is ~700KB.

    Annotations the server declares

    • destructiveHint: false
    • openWorldHint: true
    • readOnlyHint: true
    • title: List Agencies
  • search_documentsSearch Federal Register documents.Reads private dataReads text others write1 finding

    Why it is in this group

    • The name "search_documents" says "search": a word for reading.

    What it can reach

    • private-dataThe name "search_documents" says "search" and "document": it reads documents.
    • private-dataThe description's first sentence says "search" and "document": it reads documents.
    • untrusted-inputThe server marks it openWorldHint: true: it reaches things outside the server's own data.

    Findings

    • Info

      The description is 3,559 characters long, over the 2,000 this check looks for.

      Matched text

      description ends: "...reaches 5,000 documents; split the publication date range beyond that)."

      What this check looks for
    Description, as the server wrote it
    Search Federal Register documents. Finds proposed rules, final rules, notices, and presidential documents published in the Federal Register since 1994. Key parameters: - agencies: list of agency URL slugs (OR logic). Use list_agencies() to find slugs. Common: 'defense-department', 'general-services-administration', 'small-business-administration', 'defense-acquisition-regulations-system' (DFARS), 'federal-procurement-policy-office' (OFPP and the CAS Board) - far_council: True for FAR rules, proposed rules and notices (the FAR Council's output). There is no single FAR agency slug: FAR documents are filed jointly under DoD, GSA and NASA, OFPP was added only from about mid-2025 (so the OFPP slug misses earlier FAR rules and adds CAS Board documents), and 'federal-acquisition-regulation-system' holds 2 documents ever. far_council=True keeps documents filed by all three of DoD, GSA and NASA; count is exact when far_council.complete is true (narrow by date if not). Cannot be combined with agencies. For FAR clause changes only, cfr_title=48 + cfr_part='1-99' also works. - doc_types: PRORULE (proposed rule), RULE (final rule), NOTICE, PRESDOCU - term: full-text keyword search (strips stop words) - docket_id: docket identifier (token match). 'FAR Case 2023-008' = exact, 'FAR Case 2023' = all 2023 cases; partial tokens like 'FAR Case 20' match nothing - regulation_id_number: RIN (precise match, e.g., '9000-AO56') - pub_date_gte/lte: publication date range (YYYY-MM-DD) - comment_date_gte/lte: comment close date range - effective_date_gte/lte: effective date range - correction: True finds only the Office of the Federal Register's own correction notices (C1- prefix documents). Agency-issued corrections ('Final rule; correction.', 'Correcting amendment.') are ordinary RULE documents and are not flagged: search doc_types=['RULE'] with term='correction' and check each document's action. - significant: True for EO 12866 significant rules only - cfr_title + cfr_part: documents affecting a CFR location, e.g. cfr_title=48, cfr_part='52' for FAR part 52 (part accepts ranges like '1-99'; cfr_part requires cfr_title) - presidential_document_type: executive_order, proclamation, memorandum, determination, notice, presidential_order, other. Use ['executive_order'] to count or list executive orders: doc_types= ['PRESDOCU'] alone also returns proclamations, memoranda and notices. - executive_order_number: one EO by number (e.g. 14275) - order: 'executive_order_number' sorts EOs by number (pair it with presidential_document_type=['executive_order']) Presidential documents carry executive_order_number, subtype (Executive Order, Proclamation, Memorandum, ...), signing_date and presidential_document_number; other documents omit those keys. Count caps at 10,000 for broad queries (count_capped=true says so); use get_facet_counts for the true number, or narrower date ranges. per_page capped at 100 to stay within MCP response size limits. Paging: the Federal Register serves only pages 1-50 of any search (a later page silently comes back as page 1), so page > 50 is refused. total_pages is the true page count; when it passes 50, max_reachable_page and paging_note say how to reach the rest (per_page=100 reaches 5,000 documents; split the publication date range beyond that).

    Annotations the server declares

    • destructiveHint: false
    • openWorldHint: true
    • readOnlyHint: true
    • title: Search Documents

This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.

Exfil surface

Which tools read private data, which read text someone else can write, and which can send data out. One of each is a path a prompt injection can use.

48 paths in this server alone from private data, through text someone else wrote, to a way out. Denying 1 outbound tool breaks every one.

Reads private data6
Reads files, mail, messages, databases, secrets, calendars or repositories.
Reads text others write8
Brings in text someone else can write: web pages, inbound email, messages, issues and comments.
Can send data out1
Can carry data to people or systems: it sends, posts or publishes, or takes a recipient, channel or URL.
deny to break every path
{
  "permissions": {
    "deny": [
      "mcp__federal-register__open_comment_periods"
    ]
  }
}

Other servers in the same agent add their own tools to these paths. Audit your whole setup to see them together.

Remote posture

What anyone on the internet sees at the remote endpoint: an anonymous initialize and tools/list, and the public OAuth metadata.

Tools listed without sign-in

Looked at Nov 20, 2026, 19:22 UTC, unchanged since Oct 22, 2026. No credentials were sent and no tool was called. What is looked at.

Reachable
yes
HTTP status of an anonymous initialize
200
tools/list served without credentials
yes
Tools that change things, listed without credentials
1
Protocol version
2025-06-18
WWW-Authenticate on 401
not observed
resource_metadata in WWW-Authenticate
not observed
RFC 9728 protected resource metadata
not observed
Authorization servers
not observed
Legacy HTTP+SSE transport
no
Dynamic client registration
not observed
Client ID metadata documents
not observed
  • WarnThe whole server

    The endpoint served tools/list without credentials, and the list has 1 tool that changes things (1 write). Whether a call also works without credentials was not tested: tools are never called.

    Matched text

    https://federal-register.1102tools.com/mcp: tools/list returned 8 tools to an anonymous client, including open_comment_periods (write)

    What this check looks for

Findings

Matches of fixed checks, with the text they matched.

1 warn1 info
  • The server marks "open_comment_periods" read-only (readOnlyHint: true), and its own name, description or input give the class "write".

    Matched text

    annotations: {"readOnlyHint":true,"destructiveHint":false,"openWorldHint":true}; The name "open_comment_periods" says "open": a word for a change.

    What this check looks for
  • The description is 3,559 characters long, over the 2,000 this check looks for.

    Matched text

    description ends: "...reaches 5,000 documents; split the publication date range beyond that)."

    What this check looks for

History

Changes to its tool definitions and to its remote endpoint, newest first.

  1. Tools2 changed.~2 changedView diff

Install

Pinned to version 1.0.21. Values in angle brackets are yours to fill in.

claude mcp add --transport http federal-register https://federal-register.1102tools.com/mcp
claude mcp add --transport http federal-register https://federal-register.1102tools.com/mcp

Run in a terminal.

Claude Code permissions

Allow the reads, ask before a change or a way out, deny the deletes. Paste into .claude/settings.json.

.claude/settings.json
{
  "permissions": {
    "allow": [
      "mcp__federal-register__search_documents",
      "mcp__federal-register__get_document",
      "mcp__federal-register__get_documents_batch",
      "mcp__federal-register__get_facet_counts",
      "mcp__federal-register__get_public_inspection",
      "mcp__federal-register__list_agencies",
      "mcp__federal-register__far_case_history"
    ],
    "ask": [
      "mcp__federal-register__open_comment_periods"
    ],
    "deny": []
  }
}

Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. A read that can send data out (it takes a URL, say) asks first too. It assumes the server is installed as federal-register. Review it before you use it.

Capability tags behind 8 rules
  • mcp__federal-register__search_documents private-data, untrusted-input
  • mcp__federal-register__get_document private-data, untrusted-input
  • mcp__federal-register__get_documents_batch private-data, untrusted-input
  • mcp__federal-register__get_facet_counts private-data, untrusted-input
  • mcp__federal-register__get_public_inspection private-data, untrusted-input
  • mcp__federal-register__list_agencies untrusted-input
  • mcp__federal-register__open_comment_periods untrusted-input, outbound
  • mcp__federal-register__far_case_history private-data, untrusted-input

Pin

Keep the decision: fail CI when the tools change, follow every change, or show a badge.

Pin in CI

Fail a build when the tools you reviewed are not the tools the server serves today.

curl -s "https://radar.anlyon.com/api/v1/servers/com.1102tools/federal-register-mcp/verify?hash=4eee32274962639e5394038a8c54be96f8c96771b39957097e41d743fe7c41d5" | jq -e '.changed == false'

Pinned hash 4eee32274962, from Nov 9, 2026, 20:18 UTC.

Watch

Get told when its tools change: a new tool, a new description, a new class.

https://radar.anlyon.com/servers/com.1102tools/federal-register-mcp/changes.xml

An Atom feed. Works in any feed reader or Slack's RSS app.

Badge

Anlyon Radar badge for com.1102tools/federal-register-mcp
[![Anlyon Radar](https://radar.anlyon.com/badge/com.1102tools/federal-register-mcp.svg)](https://radar.anlyon.com/servers/com.1102tools/federal-register-mcp)

Source: MCP Registry · repository. Synced Oct 27, 2026, 23:17 UTC.

Federal Register MCP server: questions and answers

Answered from the scan above, not from the publisher's description.

Is the Federal Register MCP server safe?

No scan can promise that. What its definitions show: 8 tools, 1 of which can change, send or delete something, and 2 findings from fixed checks. Allow the reads, ask before each change, and deny the deletes; the permissions block on this page does that.

Can the Federal Register MCP server delete data?

No tool is classed as destructive: going by their definitions, none of them deletes, cancels or revokes anything.

Can the Federal Register MCP server do things that cannot be undone?

No tool is classed as irreversible: none of them sends, pays, publishes or merges, going by their definitions.

What tools does the Federal Register MCP server have?

8 tools: 1 write, 7 read. open_comment_periods, search_documents, get_document, get_documents_batch, get_facet_counts, get_public_inspection, list_agencies and far_case_history.

Does the Federal Register MCP server need sign-in or OAuth?

Its remote endpoint listed its tools to an anonymous client, without sign-in.

Does the Federal Register MCP server send data to outside services?

1 tool can carry data out (it sends, posts or publishes, or takes a recipient, channel or URL): open_comment_periods.

Has the Federal Register MCP server changed its tools?

Yes, once since Anlyon Radar started tracking it, most recently on Oct 23, 2026. Every change is on this page and in its Atom feed.