Skip to content
Metatext Decide — typed decisions for your agents, with a probability on every answerGet early access
Metatext
Sign Up

vexjoy-agent

Agent toolkit for Claude Code and Codex whose optional /d router uses Jev to pick agents, skills and workflows and gate dispatch on intent.

ProjectbetaMIT
View on GitHub ↗

Overview

VexJoy Agent connects plain-English requests to specialist agents, skills and workflows, with 43 agents, 62 skills, 78 hooks and 147 scripts. It works in Claude Code (/do), Codex ($do), Factory and Reasonix, and follows a route, plan, execute, verify, deliver and record pipeline.

The /d command is a Jev-powered router. Jev classifies the request, checks that the selected route preserves the requested outcome, and only then dispatches the agent, skill and pipeline. The README calls this intent preservation: /d restates the requested outcome and gates dispatch on a Jev receipt for that exact intent, to stop an agent from quietly expanding the scope. Jev is reached through Vercel AI Gateway (preferred) or TypeSafe directly. It is MIT licensed.

What the decision model decides

TypeQuestionAnswers
ChoiceWhich specialist agent, skill and workflow should handle this request?—
Yes / noDoes the selected route preserve the requested outcome?yes · no

How it works

  1. The user runs /d with a plain-English request.
  2. Jev classifies the request into agent, skill and pipeline.
  3. Jev confirms the route preserves the stated intent.
  4. Dispatch proceeds only with that receipt.

Strengths

  • Guards against scope creep with an intent check.
  • Large library of agents and skills.
  • Works across several coding agents.

Caveats

  • Jev routing is optional and needs a Gateway or TypeSafe key.
  • Large toolkit to adopt.

Getting started

export JEV_TRANSPORT=vercel and AI_GATEWAY_API_KEY, then use /d in Claude Code.

At a glance

Type
Project
Status
beta
Language
Python
Runtime
Claude Code / Codex toolkit (agents, skills, hooks)
Has code
yes
License
MIT

Use cases and integrations

How curated lists describe it

Frequently asked questions

What does /d do differently from /do?

It uses Jev to classify the request and gate dispatch on an intent-preservation check.

How is Jev reached?

Through Vercel AI Gateway (preferred) or TypeSafe directly.

Which agents does the toolkit support?

Claude Code, Codex, Factory and Reasonix.

Related projects

  • Gatekeeper

    Claude Code hook that decides which agent or skill should handle a request using YAML rulebooks plus one Jev Choice with confidence bands.

    HookbetaPython
  • jev-decision-layer

    MIT plugin and local MCP server that hands an agent's closed questions to Jev or local Laya, with policy gates and receipts.

    Agent skillproductionPython
  • agent-dispatcher

    Claude Code and Codex skill that routes a task to one of 27 specialist roles, with an optional Jev engine for catalog selection.

    ProjectbetaPython
  • Helm

    Orchestrator that probes which coding agents are installed and logged in, then has Jev pick the best one for a task.

    Agent skillbetaPython
  • jev-agent-hooks

    Hooks for Claude Code, Codex and pi that use Jev to suggest a skill each turn, plus a rule gate for subagent model routing.

    HookbetaJavaScript
  • jev-skillful

    Per-prompt capability router for coding agents that picks at most one installed skill, MCP server, subagent or command via Jev, or none.

    MCP serverbetaTypeScript