Open source · runs on Herdr · early preview

Wheems: Your own little cloud of helpers.

Wheems is an open-source framework for teams of AI agents that take on any kind of work: answering support email, replying to reviews, planning the week’s posts, looking after your SEO, or shipping code. Hand your chief of staff a job from Slack. Its crew does the work, keeps to the guardrails, and only asks you when it matters.

What your helpers can do

Not just code. Any work you can describe.

Each job is a workflow: something starts it, the crew does the steps, and anything that matters waits for your tap. Once a job has earned your trust, you can let it run by itself. To make a new one, ask the chief of staff.

  • Support inbox

    When a new email reaches support@

    Sorts each one (question, bug, refund or spam), answers questions from your knowledge base, and logs bugs for the developer.

    Waits for you on refunds, and anything it isn’t sure about

  • Reviews

    When a new review appears on Google or Trustpilot

    Drafts a reply in your brand voice, and flags unhappy customers so nobody slips through.

    Waits for you on every reply, until you’re happy to let it post

  • Social media

    When it’s Monday at 9

    Turns last week’s work and wins into posts for X and LinkedIn, checked for facts and tone.

    Waits for you on the week’s posts, before anything goes out

  • SEO and content

    When the weekly check comes round

    Checks your pages and rankings, fixes titles and descriptions, and drafts the next article.

    Waits for you on publishing

  • Bookkeeping

    When an invoice lands in your inbox

    Files it, matches it against the bank feed, and chases invoices that are overdue.

    Waits for you on any payment: the agent prepares it, you pay

  • Code

    When you ask in Slack

    Builds the fix in its own terminal, opens a pull request, and has an agent from another provider review it.

    Waits for you on the merge

Wheems is an early preview. Chatting with a chief of staff from Slack, and a knowledge base it keeps up to date, work today; email, review and social integrations come in later milestones.

What it’s made of

Five layers, one install.

You talk to a chief of staff. It hands the work to the agents you already use, unchanged, with skills, rules and a knowledge base scoped to the workspace, every one live in its own terminal on Herdr.

Works with

  • Claude Code
  • Codex
  • opencode
  • Slack
  • Telegram
  • Herdr
Wheems’s layers, from the top: you, talking from Slack or Telegram; a chief of staff for each workspace; agents on Claude Code, Codex and opencode; scoped skills, rules and the knowledge base; and Herdr, where every agent runs live in its own terminal.

How it fits together

Any job, the same seven steps.

Whatever the work, a workspace runs it the same way. Pick a job and watch it go through, or pick a step.

The opinions

Four opinions, decided for you.

You can rename or tune each one, but not remove it. That’s what makes it a framework.

  1. A 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.

  2. The 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.

  3. Scoped 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.

  4. Guardrails outside the agent

    An agent that can act can’t be stopped by asking nicely. Rules are written once and enforced where the agent can’t switch them off: the broker that holds refunds, posts and emails for your approval, spending limits, each workspace’s permissions, and GitHub for code.

Not another agent loop

Herdr is the engine. Wheems is the missing layer on top.

Chatting with one agent from Slack is useful on its own. Wheems is the set-up around it: a workspace for each part of your life, a chief of staff in each, agents from different providers, a knowledge base that keeps itself up to date, and rules the agents can’t switch off.

AspectTypical agent harnessesWheems
The agentTheir own agent loop calling a modelClaude Code, Codex and opencode, unchanged, each in its own terminal
ImprovementsArrive with the harness’s releasesEvery Claude Code or Codex update reaches every agent
Seeing the workLogs and chat repliesEvery agent is live in Herdr, where you can step in
ScopeOne instance per businessOne install, many isolated workspaces
Set-upMany choices left openOne way to do each thing
Cross-checkingOne providerAgents from different providers check each other

Getting started

From install to your first chat in Slack.

Install it once, connect Slack, create a workspace. Its chief of staff answers from your phone, and every job runs as a real agent session you can watch or take over.

Preview
$ npm install -g wheems
$ wheems setup
✓ Created ~/wheems
✓ Slack connected
$ wheems init acme
✓ Workspace acme, kept apart from the others
✓ Chief of staff on Claude Code, live in Herdr
$ wheems doctor
✓ Node.js  ✓ Herdr core  ✓ Claude Code  ✓ Slack  ✓ Service
Wheems is an early preview, and these commands may change between releases.

Helpers included.

Where it is, November 2026: version 0.0.3 is on npm, an early preview under Apache 2.0. Chatting with a chief of staff from Slack and a knowledge base it keeps up to date work today. Try it, but don’t rely on it yet: the build log says what got built each week, and why.

Or follow @mikejdeeming on X