The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local shell commands to analyze code patterns and project structure. Evidence includes the use of grep to find specific code patterns, cat and jq to inspect package.json, and npm run build to generate bundle metrics.\n- [INDIRECT_PROMPT_INJECTION]: The skill ingests and analyzes local source code, which could contain instructions designed to mislead the agent.\n
Ingestion points: Source files with .ts, .js, and .sql extensions read by grep in SKILL.md.\n
Boundary markers: No explicit delimiters are used to separate the analyzed code from the agent's instructions.\n
Capability inventory: File reading (grep, cat), directory listing (ls), and script execution (npm run build).\n
Sanitization: Content read from files is not sanitized or escaped before processing.

optimize