setup-matt-pocock-skills
Pass
Audited by Gen Agent Trust Hub on Aug 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the local repository to determine the configuration state.
- Ingestion points:
SKILL.mdinstructs the agent to read repository metadata (git remote -v,.git/config) and existing documentation files (AGENTS.md,CLAUDE.md,CONTEXT.md,CONTEXT-MAP.md). - Boundary markers: The skill does not define explicit delimiters or instructions to ignore embedded commands within the files it reads.
- Capability inventory: The skill performs filesystem writes to create configuration documents (
docs/agents/issue-tracker.md,docs/agents/domain.md,docs/agents/triage-labels.md) and modify existing agent instructions (CLAUDE.mdorAGENTS.md). - Sanitization: No explicit sanitization or validation of the ingested repository data is mentioned before it is used to generate the configuration files.
- [COMMAND_EXECUTION]: The skill uses shell commands to explore the environment and provides templates for future command execution.
- Exploration: The skill executes
git remote -vto identify the repository source and check for the presence of GitHub or GitLab remotes. - Templated commands: The output files
issue-tracker-github.mdandissue-tracker-gitlab.mdcontain instructions for using theghandglabCLIs. These include commands for creating issues, viewing comments, and managing issue dependencies (e.g.,gh api,gh issue create,glab mr list). These are legitimate development tools but represent a broad command execution surface for subsequent skill operations.
Audit Metadata