setup-matt-pocock-skills
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses shell commands to inspect the repository environment, including
git remote -vand reading.git/config. It also provides instructions for utilizing theghandglabCLI tools to perform issue management operations on GitHub and GitLab repositories. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from remote issue trackers, creating a vulnerability surface for indirect prompt injection.
- Ingestion points: Fetches issue titles, descriptions, and comments from GitHub and GitLab using
gh issue listandglab issue list. - Boundary markers: The instructions lack explicit delimitation or 'ignore instructions' warnings when handling content from external issue trackers.
- Capability inventory: The agent is empowered to write to local configuration files and execute remote tracker commands (commenting, labeling, closing issues) based on its interpretation of the environment.
- Sanitization: There is no specified mechanism for sanitizing or filtering input received from the external issue trackers before the agent acts upon it.
Audit Metadata