validate-changes
Pass
Audited by Gen Agent Trust Hub on Aug 17, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill creates an indirect prompt injection surface by interpolating raw git diff data into a prompt for another agent without sanitization or delimiters.
- Ingestion points: Staged git changes are read via
git diff --cachedand passed to theoutput-evaluatoragent in Step 3. - Boundary markers: The prompt template lacks structural delimiters (e.g., XML tags or distinct markers) to separate the untrusted diff content from the agent's instructions, potentially allowing instructions embedded in the code to influence the evaluator's verdict.
- Capability inventory: The skill uses shell execution for git operations and the Task tool for agent-to-agent delegation.
- Sanitization: No filtering, escaping, or validation is performed on the git diff output before it is processed by the evaluator.
- [COMMAND_EXECUTION]: The skill executes shell commands (
git diff --cached --stat,git diff --cached, andgit commit) to interact with the local repository environment.
Audit Metadata