Skip to content

About

Agenthood's official platform — docs, Academy, and browser-based Studio Playground for a society of AI agents with impeccable standards and zero tolerance for "fix stuff" commits.

Topics

Resources

Contributing

Security policy

Stars

1 star

Watchers

0 watching

Forks

Latest commit

 

History

420 Commits

Folders and files

NameName
Last commit message
Last commit date
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Agenthood

CI License: MIT

A full AI engineering team that earns every merge.

Live at agenthood.flabs.tech · Studio Playground · Docs


What is Agenthood?

Agenthood is a 20-member AI society — planners, builders, reviewers, auditors, and more — each a portable Markdown skill file, plus a TypeScript runtime that runs them autonomously across your whole software lifecycle.

Load them into Claude Code, Copilot, Gemini CLI, Cursor, OpenCode, or any runtime that supports skill files. Hand them a task, or let them run end to end.

No lock-in. No vendor-specific format. Setup in ~2 minutes. Every decision auditable.


Features

  • 20 specialized members — architect, reviewer, tester, auditor, security, DevOps, and more
  • 41 utility skills — specialist capability files (docker, kubernetes, jira, postgres, and more) that activate on task match
  • Delegation chains — chain agents together: tester → builder → reviewer → doorman
  • Autonomous runtime — npx agenthood run the-scribe "write a commit message"
  • Multi-agent workspaces — assemble a team, give one instruction, watch them collaborate; human checkpoints included
  • Decision intelligence — tamper-evident audit trail with hash-chain provenance
  • Conventional commit enforcement — zero tolerance for fix stuff
  • Agenthood Studio — browser-based playground with multi-agent workspaces
  • 7 LLM providers — Anthropic, OpenAI, Groq, OpenRouter, Ollama, OpenCode
  • BYOK or zero-setup — bring your own API key or use the default

Quick Links

Resource URL
Live site https://agenthood.flabs.tech
Studio Playground https://agenthood.flabs.tech/studio/playground
Studio Workspaces https://agenthood.flabs.tech/studio/workspaces
Documentation https://agenthood.flabs.tech/docs
Academy https://agenthood.flabs.tech/academy
ADRs https://agenthood.flabs.tech/adr
Releases https://agenthood.flabs.tech/releases
Upstream repo https://github.com/fworks-tech/agenthood

Project Status

Agenthood is in active development. The skill file format is stable. The TypeScript runtime is the single supported runtime (see ADR-008).

See the upstream releases for the latest changelog.


Stack

Layer Tech
Framework Next.js (App Router)
Language TypeScript
UI Mantine UI + Tailwind CSS 4
Deployment Vercel
Test runner Vitest + Playwright
LLM routing agenthood runtime (agenthood/dist/llm)

Testing

npm test          # unit tests (Vitest)
npm run test:e2e  # end-to-end tests (Playwright)
Suite Count What it covers
Unit 413 Error hierarchy, SSE parsing, secret redaction, provider routing, tool schemas
E2E 106 Config, conversations, playground, responsive layout, companion, error handling

Local development

npm install
npm run dev

Open http://localhost:3000.

The predev script runs sync-docs.mjs, sync-news.mjs, and sync-skills.mjs to fetch latest content from the agenthood repo. Requires network access to raw.githubusercontent.com.


Staging environment

The staging branch maps to the staging deployment on Vercel:

  1. Merge feature PRs to main as usual — production deploys from main.
  2. To promote a change to staging without deploying to production, merge or cherry-pick it to staging. CI and E2E run on staging pushes, same as main.
  3. In Vercel (Project Settings → Git → Production Branch), confirm main is production and add a branch alias so staging deploys get a stable URL (e.g. staging.<domain>).

Preview deployments are unchanged: every PR gets its own preview from Vercel.

Contributing

We welcome contributions! Please see CONTRIBUTING.md for guidelines.

Quick start:

  1. Fork and clone this repo
  2. npm install
  3. npm run dev — runs sync scripts automatically
  4. Create a branch: type/issue-NUMBER-short-description
  5. Follow Conventional Commits
  6. Open a PR linked to an issue

Note: Files under content/ are auto-generated from the upstream fworks-tech/agenthood repo. Don't edit them directly — edit the upstream repo or the sync scripts.


Architecture Decision Records

The full set lives in docs/adr/ — covering build-time docs sync, Studio architecture, provider routing, SSE telemetry, Turnstile gating, tool-execution replay, and more. Browse them rendered at /adr.


Related repositories

Repo Purpose
fworks-tech/agenthood The Society — 20 member skill files, 41 utility skills, TypeScript runtime, CI workflows
fworks-tech/flabs.tech Personal portfolio of the author

License

This project is licensed under the MIT License.


The Society is open to all who take the oath seriously. Membership is free. Standards are not.

About

Agenthood's official platform — docs, Academy, and browser-based Studio Playground for a society of AI agents with impeccable standards and zero tolerance for "fix stuff" commits.

Topics

Resources

Contributing

Security policy

Stars

1 star

Watchers

0 watching

Forks

Releases

Contributors

Languages