buyer-job-intent-analysis
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from various external sources, which could contain malicious instructions designed to influence the agent's behavior.
- Ingestion points:
SKILL.mddefines inputs such as user-supplied transcripts, queries, reviews, support material, and public market sources (forums, communities, RFPs). - Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded commands within the processed evidence.
- Capability inventory: The skill involves reading local project files (
icp_hypotheses.json,source_manifest.json) and writing analysis results to the filesystem (buyer_jobs.json). - Sanitization: No explicit sanitization, filtering, or escaping of the ingested external content is mentioned.
Audit Metadata