User repository on GitHub
Typed model routing for OpenClaw agents, powered by TypeSafe Jev
- Forks
- 0
- Last pushed
- Updated Oct 2026
Repo stats from the GitHub API, cached Oct 2026.
CommunityJev Agents & MCP
Jev Claw
OpenClaw plugin adding a jev_route tool that classifies task type, complexity and risk with TypeSafe Jev, then maps the result to a route through testable JavaScript policy.
- Stars
- 4
- Language
- JavaScript
- License
- MIT
- Published by
- Community
- Author
- trietphan
- Last verified
- Oct 2026
About this project
Typed model routing for OpenClaw agents, powered by TypeSafe Jev. jev-claw adds one tool — jevroute — that answers a question every multi-model agent setup runs into: Not with a prompt that asks an LLM to "think about which model is best", and not with a static if (task.includes("refactor")) rule. It classifies the task with a small typed-decision model, then applies your routing policy as ordinary code.
Quoted from the project's README. Read the full README.
Used inJev for Software & DevTools
- Languages
- JavaScript
Highlights
- jev_route returns task_type, complexity, risk, route, confidence and reasons.
- Default eight-route policy covers cheap, main, architect, debugger and reviewers.
- Eleven offline unit tests cover decide() policy without network access.
- Opt-in auto routing uses a bounded local prefilter plus one typed hook call.
Watch out
Requires OpenClaw 2026.9.0 or later with Node 24 plus a TypeSafe key, and CLI backends do not receive plugin tools. Copy reviewed against the repository README on 2026-10-04, not from running it.
Questions about this entry
Short answers drawn from the linked project or page.
Jev only classifies task_type, complexity, risk and second-opinion need. The exported decide() function maps that plus attempt history to a route, and you can fork and unit-test it offline with no key.
A hard bug starts at debugger, moves to claude-critic after two failures, and reaches frontier only with serious attempts plus failing tests. Passing tests pull it back down.
Risk is floored at high when changed_files matches sensitive paths like auth, migration, billing, secret or deploy, and high or critical risk upgrades cheap routes with a second opinion.
Automatic routing is disabled by default and qualifying requests send only fixed local labels, not free-form prompt text. Guidance mode never blocks calls; enforce mode only blocks narrow isolated debugging spawns.
OpenClaw 2026.9.0 or newer (built and checked against 2026.9.7) and Node 24+. The TypeSafe key goes in TYPESAFE_API_KEY or at ~/.config/typesafe/api_key, and the plugin installs with openclaw plugins install clawhub:trietphan/jev-claw --accept-capabilities followed by a gateway restart.
It returns task_type, complexity, risk, route, needs_second_opinion, second_opinion_route, confidence and reasons. second_opinion_route is never equal to route, and architecture decisions carry a claude-critic second opinion by default because they are the expensive ones to get wrong.
The autoRoute block defaults to guidance mode with a 2500 ms timeout, a 120000 ms cache TTL, minConfidence 0.65 and fallbackRoute main. Enabling it also requires the allowConversationAccess and allowPromptInjection hooks, and the Gateway must be restarted after the config changes.
More Jev resources like this
2GitHub stars
An MCP server packaged as both a Codex and a Claude Code plugin, exposing eight read-only tools that turn Jev probabilities into proceed/review recommendations for routing, ranking, extraction and claim checks.
Jev Agents & MCP#community#awesomejev#catalog-import
Community
E-FL/typesafe-as-a-judgeJavaScript
E-FL/typesafe-as-a-judgeJavaScript2GitHub stars
Hooks that put Jev in two spots of a coding agent's loop, picking which skill a turn needs and which model tier a subagent gets, with deterministic rule gates in front and fail-open behaviour everywhere else.
Jev Agents & MCP#community#awesomejev#catalog-import
Community
onlyjq04/jev-agent-hooksJavaScript
onlyjq04/jev-agent-hooksJavaScript7GitHub stars
A local MCP server that gives Codex a few cited excerpts instead of a whole noisy search result, with a deterministic default selector that needs no Jev account or key.
Jev Agents & MCP#community#jevbest#catalog-import
Community
suenot/codex-jev-routerJavaScript
suenot/codex-jev-routerJavaScript2GitHub stars
A Node CLI that reads a folder of Word, Excel, PowerPoint, PDF, email, CSV, HTML, code and text files, classifies each with Jev on OpenRouter, and writes a catalog, an INDEX.md and an offline report. Originals are never moved.
Jev Agents & MCP#community#awesomejev#catalog-import
Community
nexibeo/jev-organizeJavaScript
nexibeo/jev-organizeJavaScript25GitHub stars
Agent-oriented playbook for using Jev as a cheap, fast decision layer that triages, classifies, scores, and gates work before expensive browser, research, retry, or subagent steps.
Jev Agents & MCP#community#awesomejev#catalog-import
Community
Bodila51/muse-jev-playbookPython
Bodila51/muse-jev-playbookPython14GitHub stars
Opt-in Hermes plugin that routes each turn to one skill: a wide Jev Choice plus request checks gate a narrow fits round, injecting a single ignorable hint line.
Jev Agents & MCP#community#jevbest#catalog-import
From the Jev community
Posts from builders shipping with Jev right now.