revert
Pass
Audited by Gen Agent Trust Hub on Mar 26, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes multiple git commands, including
git status,git fetch,git checkout,git revert, andgit push, to automate the local and remote workflow for undoing changes in a repository. - [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection by processing external data to drive its workflow.
- Ingestion points: Metadata is retrieved from the repository using
git logand themcp__github__pull_request_readtool from the GitHub MCP server. - Boundary markers: The skill does not implement specific boundary markers or instructions to prevent the agent from obeying commands potentially embedded in PR titles or git logs.
- Capability inventory: The agent can modify files through
git revert, create new branches, and update the remote origin viagit push. - Sanitization: PR metadata is interpolated directly into the creation of new pull requests without explicit validation or sanitization steps.
Audit Metadata