commit
Pass
Audited by Gen Agent Trust Hub on Aug 25, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses local shell commands to interact with Git, including
git status,git diff,git log,git add, andgit commit. These operations are standard for the skill's purpose of repository management and are restricted to the local filesystem. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the repository's working tree via
git diff. This presents a potential surface for indirect prompt injection if the files being committed contain malicious instructions. - Ingestion points:
git diffoutput processed inSKILL.md(Step 1). - Boundary markers: The skill relies on natural language instructions for grouping but does not use explicit boundary markers for the diff content.
- Capability inventory: Access to
git addandgit commitcommands inSKILL.md(Step 7). - Sanitization: None explicitly mentioned, however, the skill mandates a manual review step in
SKILL.md(Step 6), where the user must approve the proposed commit plan before any changes are applied, effectively mitigating the risk of automated exploitation.
Audit Metadata