zaxbysauce

zaxbysauce / opencode-swarm

Public

Architect-centric agentic swarm plugin for OpenCode. Hub-and-spoke orchestration with SME consultation, code generation, and QA review.

108
9
100% credibility
Found Feb 04, 2026 at 21 stars 5x -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

OpenCode Swarm is a plugin that coordinates a team of specialized AI agents to plan, code, review, test, and maintain software projects in a structured, persistent manner.

How It Works

1
🔍 Discover the smart coding team

You find OpenCode Swarm, a helpful group of AI assistants that work together smoothly to build software features without mess or confusion.

2
📥 Add the team to your AI helper

With one easy step, you bring this team into your coding companion app so it's ready to use.

3
🧠 Pick brains for each helper

You choose which clever thinking styles each team member uses, like fast thinkers for quick scans or deep thinkers for planning.

4
💬 Chat with the Architect leader

You simply describe what you want to build, like 'add login security', and the Architect takes charge right away.

5
📋 Watch the plan come together

The team scans your project, asks experts for advice, and creates a step-by-step roadmap with checks for each part.

6
🔧 See building, checking, and testing

One by one, they code small pieces, review for mistakes and safety, write tests that pass, and update your project.

✅ Enjoy your new working feature

Your software now has the exact feature you wanted, fully tested and documented, ready to continue anytime you return.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 21 to 108 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is opencode-swarm?

opencode-swarm is a TypeScript plugin for OpenCode that runs an architect-centric agentic swarm for reliable code generation and orchestration. The architect leads a hub-and-spoke team—scanning codebases, consulting SMEs on domains like security or Kubernetes, generating code one task at a time, then routing through review and tests with QA gates. It solves multi-agent chaos by enforcing serial phases, persistent memory in .swarm/ files, and resumable sessions, so you pick up exactly where you left off.

Why is it gaining traction?

Unlike parallel agent frameworks that spawn conflicts and lose context, this delivers predictable output via phased plans, per-task reviews, and evidence bundles proving what passed QA. Developers notice the slash commands (/swarm status, plan, evidence) for inspecting progress, guardrails stopping runaway API costs, and multiple swarms for mixing cloud/local models. The verifiable execution—tests run automatically, failures fixed iteratively—builds trust for shipping working code.

Who should use this?

Backend devs building features like JWT auth in existing repos, where you need domain consultation without manual planning. Teams on OpenCode wanting AI orchestration for phased refactors, like migrating to new frameworks, with built-in SME advice and test coverage. Avoid if you're not on OpenCode or prefer single-model tools.

Verdict

Grab it if you're on OpenCode and tired of flaky agents—27 stars and 1.0% credibility reflect early stage, but 876 tests and thorough docs make it production-ready for experiments. Maturity lags big frameworks, so test on non-critical projects first.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.