The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute local Node.js scripts (precheck.mjs and cooking.mjs) found in the .agents/scripts/ directory to perform environment checks and manage task statuses. These scripts are invoked with parameters derived from user input, though the skill includes instructions for basic validation against existing directory names.
[INDIRECT_PROMPT_INJECTION]: The skill generates implementation tasks based on content read from local markdown files like spec.md and goal.md. Maliciously crafted content in these files could potentially influence the agent's behavior during the task decomposition or implementation phases.
Ingestion points: Ingests data from spec.md, goal.md, ARCHITECTURE.md, CODE-MAP.md, DEV-STANDARDS.md, PROJECT.md, and .agents/docs/ACCEPTANCE.md.
Boundary markers: None are defined; the skill does not use specific delimiters or directives to isolate content read from these files.
Capability inventory: The skill can perform file writes to create new markdown files in .agents/cooking/ and execute local shell commands via Node.js scripts.
Sanitization: There is no evidence of sanitization or filtering applied to the content retrieved from the external documentation files before it is used to generate new instructions.

to-tasks