The Agent Skills Directory

[Safe] (SAFE): No malicious patterns detected. The templates define structured workflows using internal agent slash commands.
[Indirect Prompt Injection] (SAFE): While the templates use a {{goal}} variable which could ingest untrusted data (Category 8 surface), this is a fundamental part of the developer-centric workflow. There is no automated fetching of external data from untrusted sources (e.g., web scraping) that would elevate this to a significant risk.
[Execution Safety] (SAFE): The workflows focus on standard software engineering practices like planning, testing, and reviewing. Security review steps are explicitly mentioned as context hints in review.json and coupled.json.

flow-coordinator