ci-auto-fix
Pass
Audited by Gen Agent Trust Hub on May 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill constructs shell commands using user-supplied
$ARGUMENTSand branch names retrieved from the environment (e.g.,gh pr list --head "<branch>",gh run view <run-id>). If these variables contain malicious shell metacharacters and the agent platform does not provide automated sanitization, this could lead to arbitrary command execution. - [PROMPT_INJECTION]: The skill is susceptible to Indirect Prompt Injection (Category 8) because it ingests and processes untrusted data from external sources.
- Ingestion points: Failed CI logs fetched via
gh run view --log-failed, pull request metadata fromgh pr checks, and the content of repository workflow files. - Boundary markers: The instructions lack boundary markers or explicit directives for the agent to ignore instructions that might be embedded within the logs or code files being analyzed.
- Capability inventory: The skill has significant capabilities, including the ability to modify source code, commit changes, and perform
git pushoperations. - Sanitization: No sanitization or validation steps are defined for the logs or workflow content before they are used to influence the agent's
Audit Metadata