teardown-lookup
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from the internet and local files to generate recommendations and update internal documentation, creating a surface for injection attacks.
- Ingestion points: The skill explicitly searches the web for candidate categories and practitioner sources (Step 4) and reads local user notes, documentation, and earlier AI reports (Step 5).
- Boundary markers: No explicit boundary markers, delimiters, or instructions to ignore embedded commands within the fetched external data are defined in the steps.
- Capability inventory: The skill possesses the capability to perform web searches, read local file system content (notes/docs), and write or modify local files (Step 8: "write them back as teardown's Use the map section says").
- Sanitization: There is no evidence of sanitization, validation, or escaping of the content retrieved from web sources before it is analyzed or written back to the local environment.
Audit Metadata