The Agent Skills Directory

[COMMAND_EXECUTION]: The skill directs the agent to run local CLI tools to verify project health and criteria fulfillment.
Evidence: doctrine.md identifies usage of tsc, linting tools, and project-specific scripts.
[PROMPT_INJECTION]: The skill has an indirect prompt injection surface due to reading criteria from external files.
Ingestion points: doctrine.md specifies reading from .agents/temp/repairs/.../ACCEPTANCE.md and .agents/temp/goals/.../GOAL.md.
Boundary markers: No delimiters are specified to isolate the content of these files from the agent's instructions.
Capability inventory: The agent can execute CLI commands and read workspace files.
Sanitization: No sanitization of the input file content is performed.

validate-flow