› fix signup on Safari edit src/signup/form.ts ✓ 48 tests pass $ git push origin main ✕ refused: main needs review ✓ opened PR #128
Open source · coming soon
An opinionated framework for running teams of AI agents.
Claude Code, Codex and opencode, unchanged, each live in its own terminal on Herdr. Opinr makes the set-up decisions for you, so a team of agents can work across every part of your life.
Works with
- Claude Code
- Codex
- opencode
- Slack
- Telegram
Herdr
How it fits together
One request, from your phone to the knowledge base.
Every workspace runs the same way. Watch one request go through it, or pick a step.
The opinions
Decided for you, so you don’t have to.
You can rename or tune each one, but not remove it. That’s what makes it a framework.
01A chief of staff in every workspace
Each business or area of your life is its own workspace, with its own agents, skills and knowledge. You talk to one chief of staff per workspace, from Slack or Telegram. It knows every project there, and nothing outside it.
02The knowledge loop
Every session ends with a note. Notes land in an inbox and are filed into the knowledge base for you. Agents never edit it directly, so it stays something you can trust.
03Scoped skills
Skills live in a library that’s never loaded into an agent. Each workspace picks and pins its own set, and each agent picks from that. Nothing leaks everywhere, and nothing changes under you.
04Guardrails outside the agent
Every agent has a shell, so asking it nicely enforces nothing. Rules are written once and enforced where the agent can’t switch them off: GitHub, git hooks, the broker and each workspace’s permissions.
Not another agent loop
Herdr is the engine. Opinr is the missing layer on top.
| Aspect | Typical agent harnesses | Opinr |
|---|---|---|
| The agent | Their own agent loop calling a model | Claude Code, Codex and opencode, unchanged, each in its own terminal |
| Improvements | Arrive with the harness’s releases | Every Claude Code or Codex update reaches every agent |
| Seeing the work | Logs and chat replies | Every agent is live in Herdr, where you can step in |
| Scope | One instance per business | One install, many isolated workspaces |
| Set-up | Many choices left open | One way to do each thing |
| Cross-checking | One provider | Agents from different providers check each other |
Getting started
About ten minutes to your first chat.
Install it once, connect Slack, create a workspace. Its chief of staff answers from your phone, and every conversation is a Claude Code session you can watch or take over.
$ npm install -g @opinr/hq
$ opinr setup
✓ Created ~/opinr
✓ Slack connected
$ opinr init acme --channel acme-chat
✓ Workspace acme, kept apart from the others
✓ Chief of staff on Claude Code, live in Herdr
$ opinr doctor
✓ Node.js ✓ Herdr core ✓ Claude Code ✓ Slack ✓ ServiceOpinions included.
Opinr goes open source when the first milestone works. Until then, the build log says what got built each week, and why.
Or follow @mikejdeeming on X