Skip to content
Responses API Backend API
POST /v1/responses
Bearer $PROJECT_KEY

{
  "model": "balanced",
  "user": "usr_123",
  "input": "Help me plan launch"
}
Memory on Tools governed

Agent backend for your app

Add an AI agent layer to your app with one Open Responses-compatible backend. Start with `/v1/responses`, then layer in tools, memory, approvals, observability, and channels.
API https://api.generalaugment.comAuth project-scoped keys stay server-sideRuntime managed agent turns

Start with one backend call. Keep your existing app, UX, auth, billing, and source of truth, then let General Augment provide the hosted agent layer around it.

await fetch("https://api.generalaugment.com/v1/responses", {
method: "POST",
headers: {
Authorization: `Bearer ${process.env.GENERAL_AUGMENT_PROJECT_API_KEY}`,
"Content-Type": "application/json",
},
body: JSON.stringify({
model: "balanced",
user: userId,
input: message,
metadata: { source: "app-chat" },
}),
});

When that path works, layer in tools, memory, approvals, observability, app-owned scheduled workflows, hosted channels like Telegram, WhatsApp, and SMS, or tenant-owned local connectors for private capacity.

One backend call

Call /v1/responses from your backend with a project-scoped key, stable user id, and metadata.

Memory and identity

Keep continuity per app user and link channel identities back to your own account ids.

Tools and approvals

Connect OpenAPI tools through the auth proxy and require approval for sensitive actions.

Observability built in

Track sessions, cost, tool calls, approvals, and tenant analytics from the same project surface.

  • Dashboard: https://app.generalaugment.com
  • API and webhooks: https://api.generalaugment.com
  • Docs: https://docs.generalaugment.com
  • Status: https://status.generalaugment.com; use /health/ready for hosted app polling
  • Support: support@generalaugment.com

Need the concepts first? Start with how the platform works or jump into the quickstart.