setup-matt-pocock-skills
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests repository content such as git configurations and existing documentation to suggest scaffolding options. This data is untrusted and processed without explicit boundary markers.\n
- Ingestion points: Reads repository metadata and documentation files (e.g., .git/config, CONTEXT.md) during the setup phase.\n
- Boundary markers: Not present in the skill instructions.\n
- Capability inventory: Performs file writes to the local repository and executes issue management commands via CLI tools.\n
- Sanitization: Not specified in the skill instructions.\n- [COMMAND_EXECUTION]: The skill provides instructions for the agent to use the gh and glab command-line tools for issue management, including advanced API interactions for features like issue dependencies.\n- [EXTERNAL_DOWNLOADS]: References the official GitLab CLI repository, which is a well-known and trusted service.\n- [NO_CODE]: The skill consists entirely of markdown-based instructions and configuration templates, with no executable scripts or binaries included in the skill folder.
Audit Metadata