commit
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the codebase through git diff output, which could contain malicious instructions meant to influence the agent's behavior.
- Ingestion points: The agent reads external data from the output of the git diff command to review changes (SKILL.md).
- Boundary markers: The instructions lack explicit delimiters or warnings to treat the diff content as potentially hostile data.
- Capability inventory: The skill has the capability to execute shell commands including git add and git commit (SKILL.md).
- Sanitization: No automated sanitization is performed on the diff output, although the skill requires a human confirmation step before performing any actions.
- [COMMAND_EXECUTION]: The skill uses local shell commands to interact with the version control system.
- The agent is instructed to run git status, git diff, git add, git commit, and git log (SKILL.md).
- These commands are standard for a git workflow and are executed based on the agent's internal analysis of the workspace.
Audit Metadata