The Agent Skills Directory

[SAFE]: The skill consists entirely of instructional markdown content designed to guide the agent in creating planning documents. No malicious scripts, hardcoded secrets, or unauthorized network requests were found.
[PROMPT_INJECTION]: The skill establishes a workflow where user-supplied task names are used to create files in the project root. While this is an indirect prompt injection surface, it is limited to the local creation of markdown files. Ingestion points: User-provided task names in SKILL.md. Boundary markers: Absent. Capability inventory: File-write operations (markdown plans). Sanitization: Absent.
[COMMAND_EXECUTION]: The skill references various utility scripts (e.g., api_validator.py) as examples of how a user might verify their work. The skill does not provide these scripts, nor does it attempt to install or execute them.

plan-writing