Whrit Agent Flow
whrit-agent-flow is a Claude Code plugin for the Build phase that orchestrates multi-agent Claude and Codex workflows using SPARC, swarm coordination, and MCP-style integrations.
by whrit · github.com/whrit/agent-flow
Orchestrate multi-agent Claude and Codex workflows with SPARC-style methodology, swarm coordination, and optional MCP and GitHub automation.
Add it to Claude Code
Install the plugin in Claude Code. One command, paste-ready.
/plugin install whrit-agent-flow@whrit/agent-flowBuilt to be called by your agent
Skillselion is itself an MCP server. Your agent can pull this entry and a paste-ready install config straight from the API - no copy-paste.
Retrieve this entry with skillselion.get_details("plugin:whrit/agent-flow") and the paste-ready config with skillselion.get_install_config("plugin:whrit/agent-flow").
What it does
whrit-agent-flow is a Claude Code plugin positioned as a Claude-Flow alternative that also supports OpenAI Codex, with room to add more agents later. It targets builders who outgrow single-threaded chat and want explicit orchestration: specialized agents, swarm-style coordination, SPARC methodology, test-driven workflows, and optional MCP plus GitHub automation. Solo developers running serious agent pipelines—not one prompt fixes—use it to standardize how review, implementation, and training-oriented loops run across sessions. Compared to a single SKILL.md task, this is infrastructure for how agents work together; expect advanced setup and opinionated commands. Register the whrit/agent-flow marketplace plugin in Claude Code (and configure Codex-side usage per repo docs) before relying on it for production shipping; pair it with your existing review and security practices rather than treating orchestration as a substitute for human judgment.
Highlights
- Claude-Flow-style orchestration extended to Codex (multi-agent runtime)
- SPARC methodology and TDD-oriented agent commands
- Swarm intelligence and multi-agent coordination patterns
- MCP integration and GitHub automation keywords
- Enterprise-oriented performance optimization and code-review flows
Why builders use it
Single-agent sessions bottleneck complex builds and you lack a repeatable multi-agent orchestration layer across Claude Code and Codex.
You install one orchestration plugin that coordinates specialized agents, review flows, and optional GitHub automation instead of juggling disconnected agent hacks.
At a glance
- Type - Plugin in AI Agents.
- Adoption - 0 installs, 0 stars, 0 votes.
FAQ
Who is whrit-agent-flow for?
It is for developers who want Claude-Flow-like multi-agent orchestration with Codex support and structured methodology inside Claude Code.
When should I use whrit-agent-flow?
Use it when builds need coordinated agents, code review commands, GitHub automation, or MCP-backed tooling rather than one-shot coding.
How do I add whrit-agent-flow to my agent?
Add the whrit/agent-flow plugin from its marketplace entry in Claude Code, configure orchestration commands per the README, and set up Codex according to the repo’s multi-agent instructions.
Comments
Share how you use whrit-agent-flow, gotchas, or tips for other indie builders.
No comments yet - be the first to share how you use it.