The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local Node.js scripts as part of its lifecycle.
Evidence: It runs node .agents/scripts/precheck.mjs for environment validation and node .agents/scripts/cooking.mjs status <feature> to retrieve status information.
Risk: The <feature> parameter is derived from user input or project context, which could lead to command injection if the execution environment does not properly sanitize shell arguments.
[INDIRECT_PROMPT_INJECTION]: The skill reads and processes existing goal.md and spec.md files from the project directory to perform conflict checks and updates.
Ingestion points: Processes content from .agents/cooking/<feature>/goal.md and .agents/cooking/<feature>/spec.md.
Boundary markers: Absent. The instructions do not specify the use of delimiters or 'ignore' instructions when reading these files.
Capability inventory: The skill has the authority to execute shell commands (node scripts) and perform file system operations (deleting spec.md, tasks/, and reviews/ directories).
Sanitization: Absent. There is no mention of validating or escaping the data ingested from the repository files before it influences the agent's logic or is written back to other files.

explore