interrogate
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute
git diffcommands to identify the scope of code changes for review. This is a standard and necessary operation for a development-focused code analysis tool. - [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted code diffs and intent descriptions from the user, which are then passed to subagent reviewers. There is a risk that malicious instructions embedded in the code could influence the reviewer's output.
- Ingestion points: Code diffs and intent descriptions are gathered in
SKILL.md(Step 1 and Step 2). - Boundary markers: The
references/reviewer-prompt.mdtemplate uses Markdown headers (e.g.,## Intent,## Code Under Review) to delimit external content from instructions. - Capability inventory: The skill has capabilities to read files, execute git commands, and spawn reviewer subagents.
- Sanitization: The skill does not explicitly sanitize or escape the content of the diffs or intent strings before interpolation.
Audit Metadata