The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes common bash utilities including find, awk, wc, and sed to perform structural analysis and evaluate code concentration within the local project directory. These commands are used for diagnostic purposes relevant to documentation generation.
[PROMPT_INJECTION]: The skill reads existing documentation files (AGENTS.md, CLAUDE.md) and code comments, establishing a surface for indirect prompt injection. Given the skill's primary function is to interpret codebase content, this is a known surface with limited risk. 1. Ingestion points: Existing project documentation and source code comments extracted during the discovery phase. 2. Boundary markers: No specific delimiters or instructions to ignore embedded commands are present in the sub-agent prompts. 3. Capability inventory: Local bash execution, file read/write, and parallel sub-agent task spawning. 4. Sanitization: No explicit validation or filtering of ingested file content is mentioned.
[SAFE]: No indicators of remote code execution, hardcoded credentials, or persistence mechanisms were found. The skill operates within the local environment on project-specific files.

index-knowledge