update-hr — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited update-hr (Agent Skill) and scored it 96/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 1 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 1 flagged
A bulleted imperative like {match} tells the agent to never reveal, disclose, or mention something to the user. Used adversarially it can instruct the agent to hide its tool calls or lie about what it did — stripping the transparency a user relies on to trust the agent.
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Find LinkedIn recruiter profiles for a company and add them to column E of List.xlsx.
Use WebSearch with this query:
"{company}" recruiter OR "talent acquisition" OR "hiring manager" OR "HR manager" OR "people partner" Paris OR France OR EMEA site:linkedin.com/inLocation priority: Paris first, then France, then EMEA as a fallback. Only use EMEA-level contacts when no Paris or France-specific person is found. Ignore contacts based in India, US, or APAC unless the company has no EU presence at all.
Open update_hr_contacts.py and add or extend the company entry in the HR_CONTACTS dictionary. Use the company name exactly as it appears in column A of List.xlsx. If the company is already in the dict, append the new contacts to its list — don't replace existing ones.
Run update_hr_contacts.py from the project root. The script reads existing cell content first and only appends contacts whose names aren't already there. Running it twice produces zero changes the second time.
The script reports how many rows were updated. Check that the count matches expectations.
HR_CONTACTS dict. That's the only case requiring both steps.~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.