code-review
Pass
Audited by Gen Agent Trust Hub on Aug 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill possesses a vulnerability surface for indirect prompt injection due to its reliance on processing untrusted external data.
- Ingestion points: Untrusted data enters the agent context through the output of
git diffandgit logcommands, as well as the contents of repository files located indocs/,specs/, or.scratch/directories (SKILL.md). - Boundary markers: The prompts used to spawn sub-agents lack explicit delimiters or 'ignore' instructions that would prevent the sub-agent from accidentally or maliciously executing instructions embedded within the code diffs or specification text (
SKILL.md). - Capability inventory: The skill has the capability to execute shell commands (
git) and spawn autonomous sub-agents (Agenttool), which provides an execution path for instructions derived from ingested data (SKILL.md). - Sanitization: There is no evidence of sanitization, filtering, or escaping of the content retrieved from the repository before it is interpolated into the prompts for the review sub-agents.
Audit Metadata