The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local Python utility script (scripts/sync.py) and orchestrates the execution of project-specific commands (e.g., test runners, linters, and build tools) that are dynamically discovered by probing the local repository manifests.\n- [PROMPT_INJECTION]: The skill ingests untrusted data from the analyzed repository (including package.json scripts and CLAUDE.md documentation) to populate agent instructions and execution arguments, creating a surface for indirect prompt injection should the repository content be malicious.\n- [EXTERNAL_DOWNLOADS]: The skill references the author's GitHub repository (github.com/vtmocanu/skills) for synchronization and documentation, which is categorized as a vendor resource according to the authoring context.

agent-team