qualify
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from Job Descriptions (JDs) provided as pasted text or file paths. While this is necessary for its function, the lack of boundary markers or sanitization could allow a maliciously crafted JD to influence the scoring logic or the content of the session logs written to the file system.\n
- Ingestion points: Job Description (JD) input via argument-hint as pasted text or file path.\n
- Boundary markers: Absent; there are no instructions to the agent to treat JD content as data only or to ignore embedded instructions.\n
- Capability inventory: File system read access (resume.yaml, company.yaml) and file system write access (knowledge/sessions/ path for .yaml and .md logs).\n
- Sanitization: Absent; no validation or filtering is specified for the external JD content before it is processed.
Audit Metadata