customer-research
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and synthesize information from potentially untrusted external sources, including web searches, community forums, chat history, and emails.
- Ingestion points: Untrusted data enters the agent context via Tier 3 (Team Communications) and Tier 4 (External Sources) as defined in
SKILL.md. - Boundary markers: The instructions do not define specific delimiters or "ignore embedded instructions" warnings for the processed content.
- Capability inventory: The skill is procedural; capabilities depend on the underlying agent's tools for searching and reading files/web content.
- Sanitization: No explicit sanitization, validation, or filtering of external content is requested in the methodology.
- [NO_CODE]: The skill consists entirely of markdown instructions for the AI agent and does not include any scripts, configuration files that execute code, or binary files.
Audit Metadata