The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes utility scripts (scripts/analyze_structure.sh, scripts/detect_state.sh, and scripts/estimate_tokens.sh) that use standard tools like find, grep, cat, and wc to inspect the project structure and suggest documentation placement. These scripts are intended to help the agent analyze the local development environment.
[INDIRECT_PROMPT_INJECTION]: The skill's primary function is to set up a documentation hierarchy that agents are explicitly told to read and follow as a source of truth for codebase invariants and patterns. This creates a surface where instructions embedded in project files could influence or override agent behavior.
Ingestion points: The agent is directed by the workflow in SKILL.md and references/templates.md to read AGENTS.md and CLAUDE.md files located at the project root and within subdirectories.
Boundary markers: The provided templates do not utilize security boundary markers or explicit warnings for the agent to ignore potentially malicious directives embedded in the context files.
Capability inventory: The skill includes scripts for file system traversal and reading file content. The agent typically maintains broad access to the codebase when performing these tasks.
Sanitization: There is no mechanism to validate or sanitize the natural language content within the hierarchical nodes before the agent processes them as instructions.

intent-layer