By zmuhls
Unified orchestrator for Codex CLI and AutoAgent — routes tasks to the optimal tool
Use this agent when Claude determines that AutoAgent's multi-step execution, web research, or file analysis capabilities would add value. AutoAgent excels at tasks requiring sandboxed code execution, web browsing, and iterative problem-solving.
Use this agent to intelligently route analysis tasks to either Codex (for code review, security audits, bug hunting) or AutoAgent (for multi-step coding, web research, file analysis). Spawns the appropriate sub-agent based on task characteristics.
Claude Code plugin that unifies OpenAI Codex CLI and AutoAgent into a single routing layer. Describe your task and the plugin picks the optimal tool.
/dispatch <task> — analyzes the task and routes to the best tool automatically/auto <task> — runs AutoAgent's system agents (coding, web research, file analysis)/codex <task> — runs Codex CLI for code review and static analysis (via cc-codex-plugin)The plugin also provides subagents (auto-agent, router-agent) that Claude can spawn autonomously when it detects a task would benefit from external analysis.
| Task type | Routed to | Why |
|---|---|---|
| Code review, security audit, bug hunting | Codex | focused single-pass static analysis |
| Implementation planning, edge cases | Codex | structured analytical output |
| Multi-step coding with execution | AutoAgent | Docker-sandboxed iterative coding |
| Web research, data gathering | AutoAgent | Web Surfer agent browses the web |
| File analysis (PDF, DOCX, etc.) | AutoAgent | File Surfer agent |
| Simple/trivial | Claude | no external tool needed |
brew install codex && codex login)pip install -e .)--local flag without it)# add as a marketplace source
claude plugin marketplace add zmuhls/cc-orchestrator-plugin
# install
claude plugin install cc-orchestrator-plugin@cc-orchestrator-plugin
# auto-routed — plugin picks the tool
/dispatch review auth.py for SQL injection vulnerabilities
/dispatch research the latest papers on RAG architectures
/dispatch write a CSV parser in Python
# explicit Codex
/codex find bugs in sort.py
# explicit AutoAgent
/auto --local analyze the files in the current directory
/auto --agent get_coding_agent write a web scraper for product prices
| Flag | Values | Default |
|---|---|---|
--model | any model name | COMPLETION_MODEL from env |
--agent | get_system_triage_agent, get_coding_agent, get_websurfer_agent, get_filesurfer_agent | get_system_triage_agent |
--local | (flag) | false (auto-detects Docker) |
cc-orchestrator-plugin/
.claude-plugin/
plugin.json # plugin manifest
marketplace.json # marketplace registration
scripts/
auto_oneshot.py # non-interactive AutoAgent wrapper
commands/
auto.md # /auto slash command
dispatch.md # /dispatch auto-routing command
agents/
auto-agent.md # subagent for AutoAgent delegation
router-agent.md # routing agent (Codex vs AutoAgent)
skills/
orchestrator/
SKILL.md # unified guidance for both tools
The auto_oneshot.py wrapper exists because AutoAgent's CLI (auto main) is interactive. The wrapper extracts the environment initialization logic and runs a single query non-interactively, making it compatible with Claude Code's Bash tool.
MIT
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
npx claudepluginhub zmuhls/cc-orchestrator-plugin --plugin cc-orchestrator-pluginSub-agent runner — runs agent definitions on Codex, Claude Code, Cursor CLI, or Gemini CLI from any host AI tool.
Delegate plan execution to Codex CLI via ASP. Part of cc-multi-cli-plugin. Requires the `multi` plugin.
One plugin to bridge and delegate across Claude Code, Codex CLI, and Gemini CLI — single-source AGENTS.md, shared skills, mirrored hooks and MCP servers, and full Claude↔Codex bidirectional delegation.
Complete collection of battle-tested Claude Code configs from an Anthropic hackathon winner - agents, skills, hooks, and rules evolved over 10+ months of intensive daily use
This skill should be used when the model's ROLE_TYPE is orchestrator and needs to delegate tasks to specialist sub-agents. Provides scientific delegation framework ensuring world-building context (WHERE, WHAT, WHY) while preserving agent autonomy in implementation decisions (HOW). Use when planning task delegation, structuring sub-agent prompts, or coordinating multi-agent workflows.
Cross-agent review workflow: Claude implements, Codex reviews