teach-impeccable
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill scans untrusted local project files to derive design patterns, which represents a potential surface for indirect prompt injection if those files contain malicious instructions meant to influence the agent's synthesized design context. \n
- Ingestion points: The skill explores the codebase (SKILL.md Step 1), including the README, package.json, existing components, and brand assets. \n
- Boundary markers: No explicit boundary markers or delimiters are defined for the scanning process; however, the synthesis step includes a human-in-the-loop validation via AskUserQuestion. \n
- Capability inventory: The skill writes to .impeccable.md and CLAUDE.md (SKILL.md Step 3). \n
- Sanitization: No explicit sanitization or content validation is performed on the ingested codebase data before it is synthesized into the final design context section.
Audit Metadata