The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local shell commands to inspect the repository environment and determine configuration needs. It runs git remote -v and reads the .git/config file to identify the project's host and remote origin.
[EXTERNAL_DOWNLOADS]: The skill references and provides documentation for using official command-line interfaces from well-known services. The configuration templates for GitHub and GitLab reference the gh and glab CLI tools, respectively. The issue-tracker-gitlab.md file contains a link to the official GitLab CLI repository (https://gitlab.com/gitlab-org/cli).
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it reads and processes data from local project files. 1. Ingestion points: The skill reads AGENTS.md, CLAUDE.md, CONTEXT.md, CONTEXT-MAP.md, and package.json to understand the repository's structure. 2. Boundary markers: The instructions do not specify any boundary markers or instructions to ignore embedded directives when reading these files. 3. Capability inventory: The agent is authorized to execute shell commands (git, gh, glab) and write files to the project directory. 4. Sanitization: There is no evidence of content sanitization or validation for the data ingested from the repository files.

setup-matt-pocock-skills