The Agent Skills Directory

[COMMAND_EXECUTION]: The skill analyzes the target repository by executing shell commands to extract structural and historical context.\n
Uses find to map the directory tree and identify language-specific signal files.\n
Uses git log to identify high-churn files and recent activity.- [EXTERNAL_DOWNLOADS]: When invoked in 'showboat' mode, the skill leverages the uvx package runner to fetch and execute the showboat package from the Python Package Index (PyPI).- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it processes untrusted local data.\n
Ingestion points: Reads README files, manifest files (package.json, pyproject.toml), source code, and git history across multiple files.\n
Boundary markers: Lacks explicit delimiters or instructions to ignore embedded commands within the ingested content.\n
Capability inventory: Possesses command execution capabilities via find, git, and uvx (SKILL.md).\n
Sanitization: No evidence of content validation or escaping before the data is synthesized into the final documentation.

orient