Loading ToolBench

Reading the current corpus…

mcp-govinfo MCP server · D · ToolBench

← Browse

mcp-govinfo

github.com/pipeworx-io/mcp-govinfo

GovInfo.gov MCP — full text of US government publications

2026-07-28+5 tools

58/100

Definition Quality77

Protocol Readiness35

Supportability43

Spec posture

Inferred effective spec 2026-07-28+·current baseline 2026-07-28

Static source inference · medium confidence

Current-spec patterns detected

Summary

GovInfo MCP has well-structured tool definitions with clear naming (verb_noun pattern) and reasonable descriptions (avg 150 chars). All 5 tools have input schemas with typed properties and required fields. However, parameter descriptions are sparse or missing for several tools (e.g., list_granules lacks page_size/offset_mark descriptions), and output schemas are entirely undocumented. Error handling is present but generic. No tool annotations (readOnlyHint, destructiveHint, idempotentHint) despite all tools being read-only. Missing pagination guidance in descriptions despite tools supporting it.

Tools (5)

get_granuleread onlyauthsource verified68/100

Granule metadata for a sub-unit within a package.

get_packageread onlyauthsource verified82/100

Metadata for a single package by packageId (e.g., "BILLS-118hr1234ih", "FR-2024-05-12"). Returns title, dates, citations, granule count, download links (PDF/XML/MODS).

list_collectionsread onlyauthsource verified82/100

List GovInfo collections (BILLS, CFR, USCODE, FR, CHRG, CRPT, HMAN, PLAW, SERIALSET, USCOURTS, etc.) with package counts. Use the collection code with search_packages to filter.

list_granulesread onlyauthsource verified72/100

List granules within a package — e.g., sections of a CFR title, individual entries in a Federal Register issue. Returns granule IDs + titles.

search_packagesread onlyauthsource verified80/100

Full-text + faceted search across GovInfo. Filter by collection codes (comma-separated), date range, congress, court (for USCOURTS), and free-text query. Returns package IDs and titles.

Issues (5)

HIGHtool ↗

Output schemas completely undocumented. LLMs cannot predict response structure or plan downstream tool calls. All 5 tools lack documented return types.

list_collectionssearch_packagesget_packagelist_granulesget_granule

HIGHtool-description ↗

Parameter descriptions missing or incomplete. list_granules and get_granule lack descriptions for page_size, offset_mark, granule_id. LLMs cannot infer parameter semantics from names alone.

list_granulesget_granule

MEDIUMtool ↗

No tool annotations despite all tools being read-only. Missing readOnlyHint would help agents understand safety and retry semantics.

list_collectionssearch_packagesget_packagelist_granulesget_granule

MEDIUMpaginated-result ↗

Pagination guidance absent from descriptions. search_packages and list_granules support offset_mark and page_size but descriptions do not explain pagination behavior or when to use offset_mark.

Recommendations

Document output schemas for all 5 tools. Example for list_collections: 'Returns {count: number, collections: [{code: string, name: string, package_count: number, granule_count: number}]}'.
Add descriptions to all parameters. For list_granules: 'page_size: Number of granules per page (1-100, default 100)' and 'offset_mark: Pagination cursor from previous response for fetching next page'.
Add readOnlyHint: true to all tool definitions in the McpToolDefinition interface to signal safety and retry semantics.
Enhance error messages with recovery guidance. E.g., 'GovInfo: not found (HTTP 404). Verify package_id format (e.g., BILLS-118hr1234ih) or call search_packages() to discover valid IDs.'
Add pagination guidance to search_packages and list_granules descriptions: 'Supports pagination via offset_mark cursor. Fetch next page by passing the offset_mark from the previous response.'

Score history

Overall score trend

First recorded score · v2 rubric

58/100

Scored	Grade	Overall	Spec posture	Rubric
2026-09-23	D	58	2026-07-28+	v2