The Agent Skills Directory

[SAFE]: No malicious patterns such as direct prompt injection, credential theft, or exfiltration attempts were detected.
[SAFE]: The skill does not perform any network operations or download external scripts.
[PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it processes project files (e.g., from the .tasks/ directory) which may contain untrusted data. Ingestion points: Project documentation and task files accessed via Read, Glob, and Grep. Boundary markers: None identified; instructions do not specify delimiters to isolate plan data from agent instructions. Capability inventory: The skill uses Read, Grep, Glob, Edit, and LSP tools, including file modification capabilities. Sanitization: No validation or sanitization is performed on the ingested content.

phase-review