User repository on GitHub
- Forks
- 1
- Last pushed
- Updated Sep 2026
Repo stats from the GitHub API, cached Oct 2026.
Jev directory
CommunityJev Integrations & Adapters
Jev Skill Router for Codex
A Codex plugin that routes skills with Jev: it evaluates eligible skills and passes the selected names, paths and relevance probabilities to Codex, records every decision locally, and ships a dashboard for selection rates and skill rankings.
- Stars
- 2
- Language
- JavaScript
- Reading time
- 3 min
- Published by
- Community
- Author
- Jiaqi Gu
- Last verified
- Oct 2026
Works forJev for Routing
Used inJev for Software & DevTools
- Languages
- JavaScript
- Platforms
- CLI
- Capabilities
- Open sourceRequires API key
Highlights
- Jev evaluates eligible skills and passes the selected names, paths and relevance probabilities to Codex.
- Local history records selections, no-matches, empty catalogs and failures with project, session, model, threshold and latency.
- A local dashboard shows selection rates, skill rankings and average relevance, filterable by time, project, skill or outcome, with English or Chinese UI.
- The shipped hook is bundled: plugin users do not need npm install, Python or an MCP server; app-launch startup uses an included macOS LaunchAgent.
- Requires Codex CLI with plugin hooks (developed against 0.154.0), Node.js 22+ and a TypeSafe key; defaults: threshold 0.8, maxSkills 3, port 4318.
Quickstart
bash
codex plugin marketplace add https://github.com/droid-Q/jev-skill-router.git
codex plugin add jev-skill-router@jev-skill-routerWatch out
Copy reviewed against the repo README and the madewithjev build note on 2026-10-15; the README was truncated in the settings table, so some defaults and later sections were not inspected. The README explicitly labels the dashboard screenshots as synthetic demo data, not measured Jev accuracy or real user history, and the repo lists no license in the captured metadata.
Questions about this entry
Short answers drawn from the linked project or page.
Skill routing for an agent, decided by Jev: it evaluates the eligible skills available to Codex and passes the selected names, paths and relevance probabilities along, so the choice of which skill an agent loads is a Jev decision instead of an extra LLM turn.
Jev evaluates eligible skills and returns the selected ones with relevance probabilities. The default threshold is 0.8, meaning a skill must clear that relevance probability to be selected — the setting must be greater than 0.5 and at most 1 — and maxSkills defaults to 3.
Run codex plugin marketplace add https://github.com/droid-Q/jev-skill-router.git, then codex plugin add jev-skill-router@jev-skill-router. Configure the API key, and open /hooks in the Codex CLI to review and trust the plugin's SessionStart and UserPromptSubmit commands, since installing a plugin does not automatically trust its hooks.
Selection rates, daily trends, skill rankings, average relevance and expandable records, filterable by time, project, skill or outcome, with an English and Chinese UI and light or dark appearance. The README notes the screenshots use synthetic demo data, not measured Jev accuracy or real user history.
Codex CLI with plugin hooks, SessionStart, UserPromptSubmit and app-server skills/list — developed against 0.154.0 — plus Node.js 22+ available to the process running Codex, and a TypeSafe API key with access to Jev. The desktop app and CLI must share the same local Codex configuration.
Set TYPESAFE_API_KEY in the environment that starts Codex, or create ~/.config/jev-skill-router/config.json, which the README says to keep outside repositories and restrict with chmod 600. For the desktop app, prefer the file, since apps launched from Finder may not inherit terminal environment variables.
No. The shipped hook is bundled, so plugin users do not need npm install, Python or an MCP server. Startup on app launch uses the included macOS user LaunchAgent, while session and prompt startup remain available on other platforms.
Reactions & coverage
Posts, threads, and videos about this entry from around the web.
Facebook: Intake Router keeps diagnosis out of the answer space at the front desk.
FacebookClinicnotes
X: OpenRouter Decision Model Rankings
X: herdr + jev, an agent router
X: A group chat bot with decision room routing
More Jev resources like this
A Jev-native router for Codex: before the coding model is invoked, Jev picks a native tool call from a small set of bounded reads, and Codex executes it through its own runtime and sandbox, keeping the result in the conversation.
Jev Integrations & Adapters#community#catalog-import#madewithjev
Route OpenAI Codex CLI turns through Jev: a local loopback Responses API proxy asks Jev to pick a Codex model and reasoning effort for each fresh turn, then forwards the request. If Jev is unavailable, Codex continues unchanged.
Jev Integrations & Adapters#community#catalog-import#madewithjev
Community
tiandee/codex-jev-routerJavaScript
tiandee/codex-jev-routerJavaScriptA fork of codex-with-chatgpt that adds TypeSafe Jev plus deterministic rules to decide when Codex hands planning, debugging or review to web ChatGPT and when it switches back. Smart switching is off by default, and with it off the tool behaves exactly like the original.
Jev Integrations & Adapters#community#catalog-import#logicrw
Effortlane is a local router for Codex on macOS that picks model and reasoning effort with Jev before each user turn, keeps native ChatGPT authentication, and records routing evidence locally. Shadow mode logs recommendations without changing execution.
Jev Integrations & Adapters#community#catalog-import#logicrw
An opencode plugin that turns an agent's small decisions — is this command safe, which tool fits, is this PR ready — into typed Jev questions with calibrated confidences, labelling each answer auto or escalate so the agent knows when to ask for help.
Jev Integrations & Adapters#community#catalog-import#logicrw
Community
darwintechlab/openjevJavaScript
darwintechlab/openjevJavaScriptA hook plus /jev skill for Claude Code: Jev picks the cheapest Claude tier (haiku, sonnet, opus, fable) for each prompt and estimates whether the task depends on the conversation, delegating self-contained work to a cheaper subagent.
Jev Integrations & Adapters#community#catalog-import#logicrw
From the Jev guides
Original write-ups on what this entry is for.
From the Jev AI model community
Posts from builders shipping with Jev right now.