pr-publish-agent
Pass
Audited by Gen Agent Trust Hub on Feb 21, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- Indirect Prompt Injection (LOW): The skill processes untrusted data from external sources (issue tracker) and interpolates it into command-line arguments for the GitHub CLI.
- Ingestion points: The skill reads
Parent issue ID,Parent issue URL, and/orchestra-config.json. - Boundary markers: No explicit delimiters or instructions are provided to the agent to treat input as data rather than instructions.
- Capability inventory: The skill uses
git pushandgh pr createvia subprocess calls. - Sanitization: No sanitization or validation of the
issue-idorissue-urlis performed before interpolation into theghcommand. - Command Execution (SAFE): The skill uses standard tools (
git,gh) for its primary purpose. There is no evidence of arbitrary or malicious command execution outside of its documented functionality. - Data Exposure (SAFE): While the skill reads a configuration file (
/orchestra-config.json), this is a local project configuration and does not target system-level sensitive files like SSH keys or environment secrets.
Audit Metadata