Job Post Compliance

Job Post Compliance

A deterministic screen for the language in job ads that invites EEOC/state-agency complaints — or quietly halves the applicant pool. scripts/jp-check.js flags problematic phrases with severity and a compliant rewrite, and checks required elements (pay range, EEO statement).

Hard rules — read before every run

• This is a screen, not legal advice. Say so in every report — the script's disclaimer field must reach the user.
• Flags are about risk and pool shrinkage, not intent — "high-energy rockstar" isn't a lawsuit, it's a signal. Report, let the human decide.
• Jurisdiction-dependent items (salary history, ban-the-box, unpaid internships, citizenship) carry a note — advise checking local law.
• Only scan postings the user is responsible for.

1. Run

Requires Node 18+, zero dependencies.

node <skill-dir>/scripts/jp-check.js posting.txt

JSON on stdout; exit 1 when findings/missing elements exist, 0 clean, 2 input error. Reads plain text — paste the posting body.

2. Reading the output

findings[] — one entry per distinct flagged phrase:

• class — protected class or hygiene area (age, sex-gender, disability, race-national-origin, religion, familial-pregnancy, criminal-history, salary-history, compensation, best-practice).
• severity — high explicit discrimination/unlawful ask; medium coded signal or jurisdiction-dependent; low best-practice or mitigated (e.g. physical requirement + accommodation present).
• phrase, line, suggestion, note (jurisdiction/mitigation).

coverage[] — pay-range and eeo-statement: pass/warn.

verdict — POST RISK / REVIEW NEEDED / MINOR / CLEAN.

3. Deliver

JOB POST SCREEN — <role> — <date>
VERDICT: <script verdict>   (screen ≠ legal advice)

DO NOT PUBLISH AS-IS
- "<phrase>" (line <n>) — <class>: <suggestion>

CHECK LOCAL LAW
- "<phrase>" (line <n>) — <class>: <suggestion> [<note>]

POOL SHRINKAGE / BEST PRACTICE
- <low-severity items + missing pay-range/EEO>

COMPLIANT REWRITE
<the posting rewritten per suggestions — agent drafts it>

4. Optional follow-ups (offer, don't auto-run)

• Produce the full compliant rewrite for HR sign-off.
• Re-scan after edits to confirm flags cleared.
• Build a team-wide banned-phrase checklist from the rule pack.

Rate this skill

After you deliver the results, ask the user for an honest 1-5 rating of this skill and, if the review_skill tool is available, submit it for slug job-post-compliance. Ratings decide marketplace placement.