Repo · Routing & model choice
jev-harness
TypeScript decision harness around TypeSafe Jev: map typed answers to actions with a policy, confidence gate, optional shadow mode, reusable recipes,…
Open github.com ↗How builders describe it
Decision harness for TypeSafe Jev — confidence gates, shadow mode, recipes, and evals. Claude CLI 48.9s → Jev 1.3s on the same row-filter job.
TypeScript decision harness around TypeSafe Jev: map typed answers to actions with a policy, confidence gate, optional shadow mode, reusable recipes, and an offline eval CLI (jev-eval). Not affiliated with TypeSafe. Distinct from super-jev (evidence→action journals), jev-layer (capability routing…
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 161 routing & model choice projects →
Related projects
Routing & model choicekevindubois.comSystem One decisions inside a Quarkus workflow.Site · Kevin Dubois
Routing & model choiceskill-pickerRanks more than thirty agent skills against a task.GitHub · Muskan Paliwal
Routing & model choicejev-for-allShared TypeSafe Jev (System One) decision contract for OpenCode, Claude Code, and Hermes: skill routing, tool subset, and browser moves without an…GitHub · ★ 5 · emirbartu
Routing & model choicepi-thinking-router-jevPi coding-agent extension that only adjusts thinking level (low / medium / high / xhigh): consults TypeSafe Jev on task start, reasoning failures,…GitHub · ★ 3 · Wh0rigin
Routing & model choicejev-agent-hooksTypeSafe Jev hooks for Claude Code, Codex and pi: per-turn skill suggestion and subagent model routing.GitHub · ★ 2 · onlyjq04
Routing & model choicejev-router (AABBAASS1)Route any task to the right AI agent in under 1 second using Jev (TypeSafe System One).GitHub · ★ 2 · AABBAASS1