The Agent Skills Directory

[SAFE]: No security issues detected. The skill is purely instructional, providing templates for task planning and documentation without executing code or accessing remote resources.
[PROMPT_INJECTION]: The skill transforms user-provided requirements into plans, which is a common surface for indirect prompt injection; however, no evidence of malicious instructions or safety bypasses was found in the skill's own logic. 1. Ingestion points: PRDs and task requirements mentioned in SKILL.md. 2. Boundary markers: Absent from the provided templates. 3. Capability inventory: The skill instructs the agent to generate code blocks and shell commands (e.g., bun test). 4. Sanitization: No sanitization logic is present in the instruction set.

writing-plans