The Agent Skills Directory

[COMMAND_EXECUTION]: The skill identifies and runs shell commands necessary for the development lifecycle, including starting the application via npm run dev:api, building libraries with npx nx build, and managing local database services using docker compose.
[DYNAMIC_EXECUTION]: The skill programmatically generates corrected TypeScript code and applies it to the project files using file-write operations. This behavior is the primary intended function of the autonomous debugging loop.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input in the form of application logs, stack traces, and exit codes to drive its analysis phase.
Ingestion points: Application stdout and stderr captured during the EXECUTE step in SKILL.md.
Boundary markers: None explicitly defined for log processing.
Capability inventory: execute_shell, write_file, and read_file are available to the agent following the analysis.
Sanitization: No specific sanitization of log content is mentioned before it is used to formulate code fixes.

nestjs-debug