klh-find-bugs
Pass
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted data in the form of code diffs, which creates a surface for indirect prompt injection attacks.
- Ingestion points: The skill gathers input using
git diffinSKILL.md(Phase 1). - Boundary markers: The instructions do not define clear delimiters to separate the diff data from the agent's instructions, though the task context (code review) naturally scopes the interaction.
- Capability inventory: The skill uses
gitandghtools to read data. It does not include commands for writing files, making network requests to external domains, or executing arbitrary code. - Sanitization: The skill does not implement specific sanitization of the code diffs it processes.
Audit Metadata