setup-matt-pocock-skills
Pass
Audited by Gen Agent Trust Hub on Aug 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes git, gh (GitHub CLI), and glab (GitLab CLI) to read repository metadata and manage issues.- [DATA_EXPOSURE]: The skill reads .git/config to determine repository settings, which could potentially expose sensitive authentication tokens if they are included in remote URLs.- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from repository configuration files and external issue trackers, creating a surface for embedded instructions to influence the agent. ● Ingestion points: git remote -v output, .git/config, CLAUDE.md, AGENTS.md, and issue contents retrieved via gh or glab. ● Boundary markers: Absent; there are no instructions to delimit or treat ingested content as untrusted. ● Capability inventory: The skill can write files to docs/agents/, modify project files like CLAUDE.md, and execute shell commands via gh and glab. ● Sanitization: Absent; no validation or filtering of ingested data is performed before processing.
Audit Metadata