hl-lessons
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data which could contain malicious instructions meant to influence the agent's output during analysis. \n
- Ingestion points:
SKILL.mdspecifies reading session history, PR diffs, tests, and review comments. \n - Boundary markers: The instructions lack delimiters or warnings to ignore instructions embedded in the ingested data. \n
- Capability inventory: The skill proposes modifications to linters, hooks, and CI, though it explicitly instructs the agent not to write files directly. \n
- Sanitization: There is no instruction to sanitize or validate the external content before processing.
Audit Metadata