The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external files (PRDs, Tech Specs, and task definitions) which are then used to influence the agent's code implementation and decision-making process.
Ingestion points: Steps 1.1, 1.2, and 1.3 in SKILL.md explicitly instruct the agent to read ./tasks/prd-[feature-slug]/[num]_task.md, prd.md, and techspec.md.
Boundary markers: The instructions do not define any delimiters or system-level warnings to distinguish between the skill's instructions and potentially malicious instructions embedded within the processed documentation.
Capability inventory: The skill possesses significant capabilities, including the ability to write project code, execute tests, modify the tasks.md file, and trigger other agents (task-reviewer).
Sanitization: No sanitization, validation, or filtering of the content retrieved from the documentation files is specified before the content is used to plan and execute the implementation.

executar-task