setup-matt-pocock-skills
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to use standard, well-known CLI tools (
ghandglab) for repository management. It includes explicit security warnings to prevent command injection, such as using the--body-fileflag or heredocs to avoid interpolating untrusted multiline text from issues into shell commands. - [DATA_EXFILTRATION]: During the exploration phase, the skill explicitly mandates that the agent must redact any token-bearing remote URLs or credentials before presenting findings to the user.
- [INDIRECT_PROMPT_INJECTION]: The skill establishes the infrastructure for processing issues and pull requests, which are potential vectors for indirect injection. It mitigates this risk by providing guidelines for other skills to treat issue content as untrusted, recommending permission verification for PR authors and avoiding direct execution of data-driven strings.
- [SAFE]: No patterns of obfuscation, unauthorized network exfiltration, persistence, or privilege escalation were detected. The skill's operations are confined to local configuration and authorized interaction with the repository's issue tracker.
Audit Metadata