commit-and-push
Pass
Audited by Gen Agent Trust Hub on Sep 5, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill functions by executing a variety of Git shell commands to inspect repository state, stage changes, and push to remotes.
- Evidence: Workflow steps 1, 6, 7, 8, and 9 utilize shell commands such as
git rev-parse,git status,git commit,git fetch, andgit push. - [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read and prioritize repository-specific instructions and styles, which creates an attack surface for indirect prompt injection from repository files.
- Ingestion points: The agent is explicitly told to read
CONTRIBUTING.md, recent commit messages, and other repository-specific instructions in Workflow step 2. - Boundary markers: The skill lacks explicit boundary markers or instructions for the agent to ignore potentially malicious directions found within the repository metadata or documentation.
- Capability inventory: The agent possesses the capability to execute shell commands (
git) and run "focused, relevant checks" which often involve executing repository-defined scripts (e.g., pre-commit hooks or test suites). - Sanitization: No sanitization or validation of the instructions found within repository files is specified.
Audit Metadata