test-unit
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from multiple external sources which is then used to influence the generation of test code, creating a surface for indirect prompt injection.
- Ingestion points: The skill ingests data from Sentry error logs (
sentry:search_issuesin SKILL.md), database schema information (supabase:execute_sqlin references/details.md), and web search results (firecrawl:firecrawl_searchin SKILL.md). - Boundary markers: The instructions do not specify the use of delimiters or specific warnings to the agent to ignore embedded instructions within the data retrieved from Sentry, Supabase, or Firecrawl.
- Capability inventory: The agent has the capability to write new files (unit and integration tests) and provides instructions for executing these tests via the shell (Quick Reference section in references/details.md).
- Sanitization: There is no evidence of sanitization, filtering, or escaping of the external content retrieved from these services before it is used in the prompt context to generate code.
Audit Metadata