setup-matt-pocock-skills
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides documentation templates that instruct the agent on how to use CLI tools like
ghandglabfor issue management, including API calls and shell commands. - [DATA_EXFILTRATION]: The skill instructions direct the agent to read local repository configuration files, specifically
.git/config, to identify remote repository URLs. While this file can contain sensitive data, it is used here for legitimate context discovery. - [INDIRECT_PROMPT_INJECTION]: The skill establishes a configuration surface in
docs/agents/that influences the behavior of other engineering skills. (1) Ingestion points: Reads project files includinggit remote -vand.git/config. (2) Boundary markers: None explicitly implemented in the generated markdown. (3) Capability inventory: The skill writes configuration files and updates project metadata. (4) Sanitization: None observed; the skill uses static templates.
Audit Metadata