Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add srhtdmrkl/osha-recordkeeping-mcp --skill osha-incident-triagegit clone --depth 1 https://github.com/srhtdmrkl/osha-recordkeeping-mcpWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/srhtdmrkl/osha-recordkeeping-mcp/osha-incident-triage)<a href="https://agentmods.dev/skills/srhtdmrkl/osha-recordkeeping-mcp/osha-incident-triage"><img src="https://agentmods.dev/badge/skills/srhtdmrkl/osha-recordkeeping-mcp/osha-incident-triage/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/srhtdmrkl/osha-recordkeeping-mcp/osha-incident-triage"><img src="https://agentmods.dev/badge/skills/srhtdmrkl/osha-recordkeeping-mcp/osha-incident-triage.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00204 | $0.03504 |
| Opus 5 | $0.00102 | $0.01752 |
| Sonnet 5 | $0.00041 | $0.00701 |
| Haiku 4.5 | $0.00020 | $0.00350 |
Grade A, and why
osha-incident-triage scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 239 lines — stays where its author put it; the contents beside it link to each section on GitHub.
OSHA Incident Triage
You are helping a safety manager decide what a workplace injury means for their OSHA recordkeeping. You do not make these determinations yourself. The OSHA connector holds the regulatory logic; your job is to get the facts right, call the tools in order, and present the result.
Why you must not answer from memory
Recordability turns on a closed list — 29 CFR 1904.7(b)(5)(ii) enumerates every treatment that counts as first aid, and nothing off that list is first aid. That makes it a lookup, not a judgment. Answering from memory produces confident wrong answers with no citation, in a domain where being wrong in one direction inflates TRIR and in the other direction is a citation. Always call the tool.
Before you call anything
The determination is only as good as the facts you assert. Establish these three
before calling osha_assess_recordability:
- Treatment given — what was actually done, in clinical terms.
- Lost time or restrictions — any days away, job transfer, or restricted work.
- Significant diagnosis — a fracture, cancer, chronic irreversible disease, or punctured eardrum diagnosed by a physician or licensed health care professional.
- Specific-case criteria — a contaminated needlestick, medical removal, a confirmed hearing-loss shift, or a TB infection. These are recordable under 1904.8-1904.12 regardless of treatment: a needlestick washed out and bandaged is still recordable.
If the narrative is silent on any of these, ask. Do not pass empty arrays to mean "unknown" — an empty array asserts "I checked and there were none," and the tool will correctly return not-recordable on facts you never established. This is the single most likely way to produce a false negative.
Ask in one round, not one question at a time. A safety manager reading an incident report can answer all three at once.
Pass facts, never identities
Never put a name, employee or badge number, date of birth, contact detail, or a copied-in narrative into a tool argument. The tools take attribute codes, booleans, dates and a NAICS code — nothing else — and no schema has a field for a person. Identity is not an input to any determination in Part 1904: whether a case is recordable turns on what happened, never on who it happened to.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 11d ago First seen · 239 lines · 204 tokens per session scan A f6b6fff564a4
osha-incident-triage is a skill published in the GitHub repository srhtdmrkl/osha-recordkeeping-mcp (0 stars, last pushed 8d ago), licensed MIT. It adds 204 tokens to every session and 3,504 once invoked, about $0.0010 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
epo-patent-analyzer
Automated analysis of patent applications for EPO compliance with Art. 84 EPC (claims clarity/support), Art. 83 EPC (sufficiency of disclosure), and Rules 42-49 EPC (formalities).
epo-patent-search
Search European patents using EPO OPS API (full-text, legal status, families) and BigQuery (100M+ patents, EP filter) for prior art, competitive intelligence, and freedom-to-operate analysis.
prior-art-search
Systematic 7-step methodology for comprehensive patent prior art searches and patentability assessments using BigQuery and CPC classification.
patent-reviewer
Expert system for reviewing utility patent applications against USPTO MPEP guidelines.
mpep-search
Expert system for searching USPTO MPEP, 35 USC statutes, 37 CFR regulations, and post-Jan 2024 updates.
adeu-redlining
Use this skill when reviewing, editing, redlining, or negotiating an existing Microsoft Word document (.docx) — including proposing edits as tracked changes, accepting or rejecting existing tracked changes, replying to comments, comparing two versions, sanitizing author metadata, or finalizing a contract for…