job-search
Pass
Audited by Gen Agent Trust Hub on Jun 30, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to the ingestion of untrusted data from external websites.
- Ingestion points: Job listings are extracted from hiring.cafe in
SKILL.md(Step 2) and full job descriptions are fetched from various employer career pages inSKILL.md(Step 5). - Boundary markers: The skill lacks explicit boundary markers or instructions for the evaluation subagent to disregard malicious commands embedded within the job postings.
- Capability inventory: The agent possesses the capability to write to the local file system (
DATA_DIR) and execute browser automation tools. - Sanitization: No evidence of sanitization, filtering, or escaping is found in
SKILL.mdorscripts/evaluate-jobs.mdbefore the external content is processed by the LLM.
Audit Metadata