linkedin-job-search-skill
Pass
Audited by Gen Agent Trust Hub on Sep 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data from LinkedIn job search results and user-provided job descriptions (JDs).
- Ingestion points: External data is ingested at Step 1 (Seed JD), Step 6 (LinkedIn Search Results), and Step 7 (Full Job Description pages) as described in
SKILL.md. - Boundary markers: The instructions do not define clear delimiters or "ignore embedded instruction" warnings for the external content being parsed.
- Capability inventory: The skill uses the ingested data to perform ranking and evidence extraction in
frameworks/ranking-rubric.md. - Sanitization: There is no explicit logic to sanitize or filter out potential prompt injection attacks (e.g., instructions hidden in JD text to manipulate rankings) contained within the external job data.
Audit Metadata