tomascupr

tomascupr / operant

Public

Self-hosted agents in Slack on your infra. Per-user OAuth, per-human audit, RBAC, approvals, 2,500+ SaaS tools via Pipedream Connect. MIT-licensed control plane for OpenClaw.

12
0
85% credibility
Found May 21, 2026 at 12 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
JavaScript
AI Summary

Operant is a self-hosted AI agent system for Slack that gives each employee their own AI assistant operating under their personal credentials. Unlike shared bot services, Operant tracks exactly who triggered each action, stores credentials encrypted on your own servers, and provides enterprise controls like role-based permissions, approval workflows for risky operations, and comprehensive audit logging. Employees can connect their own Gmail, Notion, GitHub, and thousands of other tools through per-user OAuth, while administrators maintain full visibility and control through a web dashboard.

How It Works

1
🔍 You discover Operant

Your company needs AI helpers in Slack but doesn't want shared bot identities or third-party data handling.

2
🖥️ You set up the control panel

You install Operant on your own servers and open the admin dashboard to get started.

3
🔗 You connect your tools

You link Slack, your AI model, and let employees connect their own Gmail, Notion, GitHub accounts through secure per-user access.

4
You set up who can do what
🛡️
Admins configure permissions

Set up roles so only the right people can access sensitive tools and data.

✅
Approvers review requests

When someone tries something risky, a designated approver gets notified to OK it.

5
💬 Your team starts chatting with AI

Employees mention the bot in Slack channels or send direct messages, and it responds using their own connected accounts.

6
📊 You monitor everything

The dashboard shows who's using what tools, how much it's costing, and every action is recorded in the audit log.

🎉 Everyone wins

Employees get helpful AI assistance while your company keeps full control, clear accountability, and complete audit trails.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 12 to 12 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 operant?

Operant is a self-hosted AI agent control plane that runs inside Slack on your own infrastructure. It sits on top of OpenClaw and adds the enterprise features that hosted agent products skip: every action lands in the audit log under the actual human who triggered it, not "the workspace." The system handles per-user OAuth for tools like Gmail, Notion, GitHub, and HubSpot through Pipedream Connect, giving each employee their own authenticated connection to 2,500+ SaaS tools. Built in JavaScript with Node 24+, it ships as Docker Compose or a Helm chart for Kubernetes deployment.

Why is it gaining traction?

The pitch cuts deep: if you use a hosted Slack agent today, every tool call runs under one shared bot identity. Your audit log shows "workspace did X" instead of "Sarah did X." Operant flips this. Each user OAuths their own credentials, the control plane enforces RBAC and approval gates, and every session, job, and tool call gets logged with the requesting human named. The six built-in roles (owner, admin, integration_admin, billing_usage_admin, member, viewer) plus arbitrary custom permissions give you the governance controls that security teams actually need.

Who should use this?

DevOps teams and platform engineers running AI coding agents in Slack who need to prove who did what. Security-conscious orgs that require per-human audit trails for compliance. Teams managing 2,500+ SaaS tool integrations without handing a third-party bot broad OAuth access to everyone's accounts. If you're already running self-hosted GitHub Actions agents or Azure DevOps agents and want the same audit rigor for your Slack-based AI tooling, this fits.

Verdict

Operant solves a real governance gap in the Slack agent space, but the numbers tell the story: 12 stars and a 0.85% credibility score suggest a project in early exploration rather than production maturity. The docs are thorough and the architecture is sound, but the star count means limited community validation. Worth evaluating for serious compliance needs, but treat it as an early-stage project that needs proving in real deployments before trusting it with sensitive credentials.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.