CalibratedDecisions.

Repo · Routing & model choice

claude-model-guard

Claude Code hooks plus a CLI picker where TypeSafe Jev chooses the model tier (Haiku, Sonnet, Opus or a long-form tier) for every prompt and can…

Open github.com ↗

From the repository

Pick, and optionally enforce, the right Claude Code model tier per prompt using Jev (a cheap classifier model) instead of guessing yourself.

How builders describe it

Claude Code hooks plus a CLI picker where TypeSafe Jev chooses the model tier (Haiku, Sonnet, Opus or a long-form tier) for every prompt and can block a prompt sent to the wrong model when it confidently disagrees; tier criteria are editable and there are no dependencies beyond Node's fetch.

The decision Jev makes

Pick an agent, a model, a skill from a catalog, or a function and its arguments.

Where it fits

Deciding which agent, model, skill or function should handle a request. Jev is a natural router: the options are fixed and the confidence tells you when to fall back. All 348 routing & model choice projects →

Related projects