The Agent Skills Directory

[SAFE]: No malicious commands, obfuscation, or unauthorized data access patterns were detected. The skill's primary focus is on project management, implementation discipline, and quality control.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to read and execute tasks from external plan files (e.g., 03-plan.md). While this constitutes an indirect prompt injection surface where instructions in the data could influence the agent, the skill mitigates this through structured task contracts and this behavior is a core part of its primary intended purpose.
Ingestion points: The skill reads docs/features/<name>/03-plan.md and docs/features/<name>/plans/*.md to extract implementation tasks and verification commands.
Boundary markers: The skill relies on specific markdown formatting (e.g., ### Task N) to delimit instructions within the ingested files.
Capability inventory: The skill allows for file modification, test execution (via build-quality-tdd), and sub-agent dispatching.
Sanitization: No specific content sanitization or escaping of the plan data is mentioned, as the data is expected to be authored by the user or the agent's planning phase.

build-workflow-execute