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

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.

ProjectbetaMIT
View on GitHub ↗

Overview

Agent Dispatcher is a retrieval and context helper for Claude Code and Codex. It finds the code a task needs using symbols, source text, code relationships and Git history, assembles relevant excerpts, pairs the work with specialist guidance and makes verification explicit. It ships 27 roles, 79 local skills, references to external skills, recipes and MCP servers.

The list summary describes it as routing a task to one of the 27 roles and defining what evidence counts as done. The README mentions an optional Jev decision engine, off in all scopes by default, that can delegate catalog selection to an external API. Normal use needs no extra API key.

What the decision model decides

TypeQuestionAnswers
ChoiceWhich specialist role (and skills) fits this task?(inferred)27 roles, e.g. reviewer · debugger

How it works

  1. User invokes the dispatcher with a task
  2. Dispatcher gathers code context from symbols, text and Git history
  3. A role is selected from the catalog (optionally via Jev)
  4. Work proceeds with role guidance and explicit verification evidence

Strengths

  • No extra key for normal use
  • Jev selection is opt-in

Caveats

  • Jev is a minor, disabled-by-default part
  • No native Windows install

Getting started

claude plugin marketplace add nahid-sparktales/agent-dispatcher && claude plugin install agent-dispatcher@agent-dispatcher (Codex: python3 install_codex.py)

At a glance

Type
Project
Status
beta
Language
Python
Runtime
Claude Code plugin / Codex skill
Has code
yes
License
MIT

Use cases and integrations

How curated lists describe it

  • Routes a Claude Code or Codex task to one of 27 specialist roles and defines what evidence will count as done.

    MrJev/awesome-jev

Frequently asked questions

Is Jev required?

No, the Jev decision engine is off by default and optional.

What does it route?

Tasks to one of 27 specialist roles with matching skills.

Which agents are supported?

Claude Code (as a plugin) and Codex (as a skill).

Related projects

  • 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
  • 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
  • matchcn

    MCP server and semantic index that tags shadcn-format components once and matches a plain-language brief to them deterministically.

    MCP serverbetaTypeScript
  • 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.

    ProjectbetaPython