review
Pass
Audited by Gen Agent Trust Hub on Oct 19, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes
gitandgh(GitHub CLI) to retrieve pull request information and code diffs. These tools are used for their intended purpose of repository analysis. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted data from external sources, such as pull request diffs and branch content, which could contain instructions intended to influence the subagents.
- Ingestion points: Untrusted diff data is ingested in
SKILL.mdusinggh pr diffandgit diffcommands. - Boundary markers: The skill implements reasoning isolation by running each reviewer in its own subagent and mentions using worktree isolation for audits that touch files.
- Capability inventory: The skill in
SKILL.mdhas the capability to execute shell commands, spawn subagents, and call external audit skills. - Sanitization: The skill enforces a strict "Read-only" policy, explicitly forbidding any modifications, staging, or commits to the repository.
Audit Metadata