MCP server exposing 5 tools for osha.
This URL is a JSON-RPC 2.0 endpoint over HTTP. Issue POST requests with a JSON-RPC body. Browsers and search crawlers land here on GET.
POST https://gateway.pipeworx.io/osha/mcp
Content-Type: application/json
{"jsonrpc":"2.0","id":1,"method":"tools/list"}osha_standard — Get the full text of one OSHA standard (US workplace-safety / occupational safety and health regulation) by its citation. OSHA standards are 29 CFR — this returns the exact regulatory wording currently in force. Answers "what is the OSHA standard for X", "does OSHA require X", "read the text of 1910.132", "the OSHA fall protection rule". Forgiving citation input: "1910.132", "29 CFR 1910.132", "§1910.132", "1926.501", even "1910.132(a)" (paragraph is stripped to the section). Covers 29 CFR 1910 general industry, 1926 construction, 1915/1917/1918 maritime, 1928 agriculture, plus recordkeeping (1904) and PPE / hazard communication / lockout tagout / respiratory protection / confined space / bloodborne pathogens standards. Pass a whole part (e.g. "1910" or "1926") to get that part's section list instead. Example: osha_standard({ citation: "1910.132" }) → PPE general requirements; osha_standard({ citation: "1926.501" }) → construction fall protection. Keyless.osha_search — Keyword search across OSHA standards — the US workplace-safety / occupational safety and health regulations in 29 CFR (parts 1900–1990). Answers "what OSHA standards cover X", "OSHA regulation / workplace safety requirement for X", "find the OSHA rule about X". Great for topics: fall protection, hazard communication (HazCom / GHS), lockout tagout (LOTO), respiratory protection, personal protective equipment (PPE), confined space, bloodborne pathogens, machine guarding, scaffolding, excavation, permissible exposure limits, recordkeeping. Returns matching OSHA standards with citation, heading, excerpt, and source URL. Results are filtered to OSHA parts only (1910 general industry, 1926 construction, 1915/1917/1918 maritime, 1928 agriculture, 1904 recordkeeping, etc.) — Title-29 wage/hour parts are excluded. Example: osha_search({ query: "fall protection" }); osha_search({ query: "lockout tagout", limit: 15 }). Keyless.osha_inspections — OSHA inspections for a named establishment and/or state — the enforcement record OSHA actually took, not the standards. Answers "has OSHA inspected X", "list OSHA inspections for X", "OSHA inspections in state X since date Y". Returns each inspection's address, NAICS code, whether it was a safety or health inspection, the inspection type (Accident / Complaint / Programmed / Referral / Fatality-Catastrophe / etc.), scope, open/close dates, and open-vs-closed status. Establishment matching is a substring match against OSHA's name on file (case-insensitive on input, matched upper-case) — a shorter fragment of the name finds more. Requires establishment or state (or both); `since` narrows to inspections opened after that date. Example: osha_inspections({ establishment: "Walmart", state: "TX" }); osha_inspections({ establishment: "Moran Towing", since: "2020-01-01" }). Live US Dept. of Labor data, refreshed daily. Use osha_inspection with the returned activity_nr for violations and penalties.osha_inspection — One OSHA inspection by its activity number (from osha_inspections or osha_establishment_summary): the inspection record plus every violation cited — standard cited, violation type (Serious/Willful/Repeat/Other-than-Serious), issuance and abatement dates, instances, employees exposed, and BOTH the initial and current (post-contest/settlement) penalty amounts, plus the summed totals. Answers "what violations came out of inspection X", "what did OSHA fine them for this inspection", "penalty amounts for activity number X". Example: osha_inspection({ activity_nr: 348458035 }). Live US Dept. of Labor data.osha_establishment_summary — Rolled-up OSHA enforcement history for an establishment: how many inspections, how many violations, total initial and current penalty amounts across all of them, the most recent inspection date, and the distinct establishment names OSHA has on file matching the search (chains and renamed sites often appear under several). Answers "has OSHA fined X", "total OSHA penalties for X", "how many times has OSHA inspected X", "OSHA enforcement summary for X". Narrower and faster to read than paging through osha_inspections yourself. Example: osha_establishment_summary({ name: "Walmart", state: "AR" }). Live US Dept. of Labor data, refreshed daily; capped at the 50 most recent inspections per establishment — narrow with state for a complete count on a national chain.Code samples (curl / TypeScript / one-click client install), schemas, and the live playground are on the pack page:
https://pipeworx.io/packs/osha/
Pipeworx is an open MCP gateway connecting AI agents to live data. pipeworx.io