vn-job-search
Pass
Audited by Gen Agent Trust Hub on Nov 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill evaluates external job descriptions to determine application fit, creating a surface for indirect prompt injection.
- Ingestion points: Untrusted content is ingested from job descriptions via the
get_jobtool and through direct web page fetching in Step 3. - Boundary markers: The prompt instructions do not specify any delimiters or instructions for the agent to ignore or isolate potential commands embedded within external job descriptions.
- Capability inventory: The skill allows the agent to execute actions with real-world impact, such as
request_referral,create_cv_variant, andtrack_job, which are triggered based on the results of the automated evaluation. - Sanitization: No sanitization, filtering, or validation is implemented to verify the integrity of job description content before it is used to influence the decision-making logic.
Audit Metadata