The Agent Skills Directory

[SAFE]: The skill is a procedural guide that establishes a 'hard gate' for software development workflows. It mandates a design-first approach, requiring user approval of architectural plans before executing any code-related tasks.
[INDIRECT_PROMPT_INJECTION]: The skill's workflow involves reading external data from the project environment, which presents a standard surface for indirect prompt injection.
Ingestion points: The skill instructions direct the agent to 'explore project context' by checking local files, documentation, and recent commit records.
Boundary markers: There are no defined delimiters or specific instructions to isolate content read from files from the agent's internal logic.
Capability inventory: The skill has the capability to write markdown files to the local filesystem (under docs/plans/) and invoke downstream implementation skills.
Sanitization: The skill does not include steps to sanitize or validate the content retrieved from the project context before incorporating it into the design process.

brainstorming