The Agent Skills Directory

COMMAND_EXECUTION (SAFE): The skill uses the Bash tool and lifecycle hooks to execute local commands such as npm run lint, which is consistent with its primary purpose as a coder agent.\n- PROMPT_INJECTION (LOW): The agent processes external data from tasks and files which could contain malicious instructions (indirect prompt injection surface).\n
Ingestion points: The $TASK environment variable and local files accessed via Read and Grep tools.\n
Boundary markers: Absent; the skill does not use delimiters to isolate instructions from data.\n
Capability inventory: Includes Bash for command execution and Write/Edit for file modification.\n
Sanitization: No explicit sanitization or validation of input data is defined in the skill logic.

core-coder