The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes standard search and inspection commands (rg, cat) to analyze the repository's technology stack by looking for specific file extensions and manifest files (e.g., package.json, Cargo.toml). This is a benign and expected use of tooling for repository analysis.
[INDIRECT_PROMPT_INJECTION]: The skill has a data ingestion surface as it reads and processes repository files (package managers, lockfiles, and documentation) to build an evidence table. While it lacks explicit boundary markers for this data, the risk is minimal as the ingested content is used for classification purposes and not directly executed as high-privilege instructions. The capability inventory includes file reading and providing a structured install plan, but no direct network exfiltration or unsafe dynamic execution patterns were detected.

agent-sort