typesafe-ai/skills on GitHub
- Stars
- 2,037
- Forks
- 110
- License
- MIT
- Last push
- 12 Sep 2026
Read from GitHub on . Counts change daily.
This is TypeSafe’s own agent skill, published from the typesafe-ai GitHub organization. It installs into Claude Code as a plugin through the marketplace, or into other agents through skills.sh, which asks which agent you use and installs the skill into the project by default; the -g flag installs it globally.
Once installed, the skill helps an agent design a TypeSafe workflow and compose typed judgments in code. It tells the agent to treat the live TypeSafe docs as the source of truth and read them during the task, starting from the documentation index and the closest cookbook. It also maps question shapes to primitives: choice for picking one of a defined set, noul for whether a condition holds, and score for a position along a described dimension. The README’s example prompt asks the agent to route incoming support tickets by department, with human review for uncertain decisions. In Claude Code you can also call the skill directly with /typesafe:typesafe-ai.
