Build AI teammates that watch and act

Open-source infrastructure for AI teammates that watch, remember, and act. Connectors and webhooks build a live org knowledge graph; agents look it up and branch into a sandbox to do work.

Paste into your coding agent to scaffold a project.

Or start it yourself:

1

Connect your data

Pick the systems it can read. Lobu turns those updates into live customer memory.

HubSpotStripeZendeskGmailGoogle DriveGitHubNotionSnowflakePostgreSQL50+ more
Use existing connectors, or let your agent write code to connect any data.
Watching changes across every sourceConnector SDK
2

Define the goal

Tell it what to watch for and when to ask before acting.

“Watch every account for churn risk. If renewal is within 30 days and health drops, draft a CSM check-in for approval.”

Runs every weekday · 9:00Watchers
3

Lobu works autonomously

It scans memory on schedule, spots the account at risk, and keeps the evidence attached.

Acme Corp
Jordan Lee, VP Eng
at risk
plan Enterprise · 120 seatsrenewal Jun 30 · 21 days
Derived from recent events:
"API latency on /sync" ticket opened
2d ago · Zendesk
Logins fell 38% over 14 days
5d ago · Product
Renewal date set to Jun 30
9d ago · CRM
Globex Inchealthy
+ 17 more customersMemory & entities
4

You review and approve

You can edit the draft, send it, or leave it.

R
Revenue agent
online
Heads up: Acme Corp is trending toward churn. Logins are down 38% over 14 days and their renewal is in 21 days. Want me to draft a check-in for their CSM?12:01
Yes, and include the usage drop.12:01
Drafted. Saved it to the Acme account and pinged @dana.12:01
Replies in your team's chat, or your own app over the API:
See source
For engineers

The whole agent, in code.

One project defines it end to end: the agent, its connectors, the memory schema, watchers, and skills. Write it yourself, or let your coding agent generate it. Pick a piece to read the code.

lobu.config.tslobu.config.ts
import {
  connectorFromFile,
  defineAgent,
  defineConfig,
  defineEntityType,
  defineRelationshipType,
  defineWatcher,
  reactionFromFile,
  secret,
} from "@lobu/cli/config";
import type SalesforcePipelineConnector from "./salesforce-pipeline.connector.ts";
import type accountHealthMonitorReaction from "./account-health-monitor.reaction.ts";

const sales = defineAgent({
  id: "sales",
  name: "sales",
  description:
    "Help revenue teams track account health, rollout progress, and renewal signals",
  dir: ".",
  providers: [
    {
      id: "anthropic",
      model: "claude/sonnet-4-5",
      key: secret("ANTHROPIC_API_KEY"),
    },
  ],
  network: {
    allowed: [
      "github.com",
      ".github.com",
      ".githubusercontent.com",
      "registry.npmjs.org",
      ".npmjs.org",
    ],
  },
});

// entity types and relationships defined here…

const accountHealthMonitor = defineWatcher({
  agent: sales,
  slug: "account-health-monitor",
  name: "Account health monitor",
  schedule: "0 */12 * * *",
  notification: { priority: "high", channel: "both" },
  tags: ["sales", "health", "renewals"],
  minCooldownSeconds: 1800,
  reaction: reactionFromFile<typeof accountHealthMonitorReaction>(
    "./account-health-monitor.reaction.ts"
  ),
  prompt:
    "Poll CRM data for tracked accounts. Track expansion progress, risk level changes, and renewal timeline.\n",
});

export default defineConfig({
  connectors: [
    connectorFromFile<typeof SalesforcePipelineConnector>(
      "./salesforce-pipeline.connector.ts"
    ),
  ],
  org: "sales",
  orgName: "Sales",
  orgDescription:
    "Help revenue teams track account health, rollout progress, and renewal signals",
  agents: [sales],
  entities: [organization, product, region, renewalRisk, team],
  relationships: [affects, expandedInto, runs],
  watchers: [accountHealthMonitor],
});
Examples

Explore agent workflows.

Each example shows the sources, memory, and actions for one AI teammate.

Run anywhere

Local, self-hosted, or managed.

Local

Run on your laptop.

Boot the gateway, workers, memory, and embeddings with one command.

Self-host

Run in your cloud.

Docker, cloud VM, or Kubernetes when data and controls need to stay with you.

Lobu Cloud

Let Lobu run it.

Use the same project with managed isolation, secrets, and upgrades.

Build your first
multi-user agent.