pushing-commits-to-the-repo
Pass
Audited by Gen Agent Trust Hub on Aug 14, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read and triage pull request comments from both humans and bots. This ingestion path is vulnerable to indirect prompt injection, where an attacker could embed malicious instructions in a comment to influence the agent's actions during the PR lifecycle.
- Ingestion points: PR comments are processed as part of the 'After you push — the loop' section (SKILL.md).
- Boundary markers: No explicit delimiters or instructions to ignore embedded commands within comments are provided.
- Capability inventory: The agent can edit PR labels (
gh pr edit), commit code, and push to remote branches. - Sanitization: No sanitization or validation of comment content is described.
- [COMMAND_EXECUTION]: The skill uses standard command-line tools for development workflows. These include
gh(GitHub CLI) for managing labels and pull requests,gitfor commit management, and a localpre-push-reviewscript for quality checks. These tools are used for their intended purpose in a repository maintenance context.
Audit Metadata