go-to-market-strategy
Pass
Audited by Gen Agent Trust Hub on Sep 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from local workspace files to inform its strategy generation, which represents a potential injection surface if those files contain untrusted content.
- Ingestion points: Processes sections 2 through 5 of
/foundation/brain.mdand reads patterns from/context/meta-patterns.mdas described inSKILL.md(Step 0). - Boundary markers: The instructions do not specify explicit delimiters or "ignore instructions" wrappers for the ingested brain content.
- Capability inventory: The skill is restricted to reading local markdown files and appending log entries to
/context/skill-sessions.md(Step 5). It does not have network access or arbitrary command execution capabilities. - Sanitization: No explicit sanitization or validation logic is defined for the content read from the brain or meta-patterns files.
Audit Metadata