# Build on an AI platform you own

*Source: https://ibl.ai/developer*

For Developers

**ibl.ai is the agentic AI platform where you own all the code and the data. Its 5 repositories are public on GitHub — the application layer, the SDK, 157 agent skills and two MCP servers, and the infrastructure CLI — so you can read the source, scaffold an app, and run the whole stack on your own cloud, on-premise, or fully air-gapped, model-agnostic across any LLM and with no per-seat pricing.**

Not an API you rent — a full platform you keep. 1.6M+ users across 400+ organizations run the platform this way, including NVIDIA, MIT, and Syracuse University.

- Browse 161 skills: /developer/skills
- Explore the source on GitHub: https://github.com/iblai
- Read the API reference: /developer/api-reference

161 skills · iblai/base as of Nov 18, 2026 (commit b7569f8, https://github.com/iblai/base/commit/b7569f8c8cc7ac0d269238eb936196359fdf0e86)

## Where do I start on ibl.ai?

Three doors.

Developers come to ibl.ai to do one of three things. Each door below is one command and the three pages that take you from there to running code — every one generated from the repository it documents, so it cannot drift.

### Build an agent

Create an agent, give it a prompt, tools, memory and a dataset, test it against a benchmark, and talk to it from any MCP client — with a screen or headless over REST.

```
npx skills add iblai/base --all
```

from github.com/iblai/base: https://github.com/iblai/base

- Agent skills — create, chat, tools, memory, evals: /developer/skills#agents
- Agents API quickstart: /developer/platform/agents/quickstart
- Run agents on your own server (OpenClaw): /developer/guides/claw-setup
- Connect a data source — MCP & agent skills: /developer/connectors

### Build an app

Scaffold a production Next.js app with SSO, agent chat, profiles, notifications and analytics wired to the platform, then add features one skill at a time.

```
/iblai-vibe-start
```

from github.com/iblai/base: https://github.com/iblai/base

- Getting started — the whole journey: /developer/base/getting-started
- Every skill, searchable: /developer/skills
- The domain model most apps are built on: /developer/base/domain-model

### Run the platform yourself

Provision the whole stack on your own AWS or GCP account with one interactive CLI, or bootstrap an existing server — the CLI and its docs live in github.com/iblai/infra-cli.

```
pip install ibl-cli && ibl setup
```

from github.com/iblai/infra-cli: https://github.com/iblai/infra-cli

- Infrastructure CLI — README and AWS setup: /developer/infra-cli
- Google Cloud (GCP): /developer/infra-cli/gcp
- Source: github.com/iblai/infra-cli: https://github.com/iblai/infra-cli

## Which skill builds the part of the agent I need?

161 skills, one install.

Every skill is a how-to a coding agent runs and a human can read. Search by what you are trying to do, or start from the parts of an agent below — each has a skill with a screen (`iblai-vibe-*`) and a headless twin over REST (`iblai-api-*`).

| Part of the agent | With a screen | Headless |
|---|---|---|
| Create and configure an agent | /iblai-vibe-agent-create (/developer/skills/iblai-vibe-agent-create) | /iblai-api-agent-create (/developer/skills/iblai-api-agent-create) |
| Chat with it (streamed, with tools) | /iblai-vibe-agent-chat (/developer/skills/iblai-vibe-agent-chat) | /iblai-api-agent-chat (/developer/skills/iblai-api-agent-chat) |
| Give it tools and MCP servers | /iblai-vibe-agent-tool (/developer/skills/iblai-vibe-agent-tool) | /iblai-api-agent-mcp (/developer/skills/iblai-api-agent-mcp) |
| Memory | /iblai-vibe-agent-memory (/developer/skills/iblai-vibe-agent-memory) | /iblai-api-agent-memory (/developer/skills/iblai-api-agent-memory) |
| Documents and retrieval | /iblai-vibe-agent-dataset (/developer/skills/iblai-vibe-agent-dataset) | /iblai-api-agent-dataset (/developer/skills/iblai-api-agent-dataset) |
| Test it before users do | /iblai-vibe-agent-evals (/developer/skills/iblai-vibe-agent-evals) | /iblai-api-agent-eval (/developer/skills/iblai-api-agent-eval) |
| Code execution in a sandbox | /iblai-vibe-agent-sandbox (/developer/skills/iblai-vibe-agent-sandbox) | /iblai-api-agent-sandbox (/developer/skills/iblai-api-agent-sandbox) |
| Embed it anywhere | /iblai-vibe-agent-embed (/developer/skills/iblai-vibe-agent-embed) | /iblai-api-agent-embed (/developer/skills/iblai-api-agent-embed) |

- All 161 skills by folder and kind: /developer/skills
- What a skill kind means: /developer/base/skill-kinds

## How do I set up my coding agent for ibl.ai in one step?

Written for agents.

1. Install every ibl.ai skill for Claude Code, Cursor and Codex — the same command refreshes them later.

```
npx skills add iblai/base --all
```

2. Type the first skill in your coding agent; it asks what you are building, connects your organization and writes your credentials to .env.

```
/iblai-vibe-start
```

3. Give the assistant the SDK's components, hooks and page templates while it writes your app. Details: /developer/base/mcp-servers

```
npx -y @iblai/mcp
```

4. Point any MCP client at a deployed agent for a live conversation — streamed responses, tools, retrieval. Details: /developer/skills/iblai-api-agent-chat

5. Everything on this site is readable by an agent: llms.txt, a Markdown twin of every page, and the agent-skills discovery index. Details: /llms.txt

## Which ibl.ai repositories can I read and run myself?

Highlighted repositories.

These 5 highlighted repositories are the heart of the platform — not every public repo; github.com/iblai (https://github.com/iblai) lists them all. Most AI vendors give you an endpoint; ibl.ai gives you the source you can read, fork, run, and extend. Every card below shows the live release straight from GitHub.

Each card shows the repository's latest release tag and release date, the date of its last update, its primary language and its star count, fetched live from GitHub.

### iblai/os — Application layer

The open-source platform itself. Create and deploy customizable AI agents with multi-LLM support, real-time chat and voice, document training, analytics, user management, SSO, and embeddable widgets.

TypeScript. Powers Agentic OS (/product/agentic-os).

- Read the documentation: /developer/os
- Source: https://github.com/iblai/os

### iblai/base — Build

Scaffold a production-ready Next.js app on ibl.ai — SSO auth, AI chat, profiles, notifications, and analytics wired up out of the box, with pre-built components and Claude Code skills.

JavaScript. Powers Agentic Base (/product/agentic-base).

Command: `npx skills add iblai/base`

- Read the documentation: /developer/base
- Source: https://github.com/iblai/base

### iblai/infra-cli — Deploy

An interactive CLI that provisions the whole platform on your own AWS or GCP account with Terraform and configures servers with Ansible — and it can bootstrap any existing server.

Python.

- Read the documentation: /developer/infra-cli
- Source: https://github.com/iblai/infra-cli

### iblai/docs — Reference

The documentation source — every page published at ibl.ai/docs, versioned in the open and available for contributions.

Markdown.

- Read the platform docs: /docs/os/overview
- Source: https://github.com/iblai/docs

### iblai/lms — Application layer

Skills-intelligence platform on the same SDK — discover courses, track competencies, and earn credentials. A second complete application you can read end to end.

TypeScript. Powers Agentic LMS (/product/agentic-lms).

- Read the documentation: /developer/lms
- Source: https://github.com/iblai/lms

## How do I deploy and run ibl.ai on my own infrastructure?

How-to, by goal.

Cloud, VPC, on-premise or fully air-gapped — each answer below is a page generated from the repository that ships the code, so the steps are the ones the engineers run.

- How do I deploy ibl.ai on AWS? /developer/infra-cli
- How do I deploy ibl.ai on Google Cloud? /developer/infra-cli/gcp
- How is the infrastructure laid out? /developer/infra-cli/architecture
- How do I run the OS app locally? /developer/os/running-the-app
- How do I host the OS app as static files? /developer/os/deployment
- How do I run the OS app against my own platform deployment? /developer/os/platform-deployment
- How do I build the macOS desktop app? /developer/os/macos-builds
- How do I build the iOS app with Tauri? /developer/os/tauri-ios-setup
- How do I set up an OpenClaw agent server? /developer/guides/openclaw-server-setup
- How do I set up a NemoClaw server with guardrails? /developer/guides/nemoclaw-server-setup
- What is the ontology layer, and how does it make my systems queryable? /developer/guides/ontology
- How do I deploy the ontology layer on-premise? /developer/guides/ontology-install
- What is the security model of an app built on ibl.ai? /developer/base/security-model

## How does ibl.ai compare with other developer platforms?

Compare.

The one axis where every hosted platform is identical and ibl.ai is the opposite: you receive the source code and keep the data. Each comparison below leads with that and then goes feature by feature.

- ibl.ai vs LangGraph: agent framework or platform: /resources/comparisons/self-hosted-ai-vs-langgraph
- ibl.ai vs OpenAI Agents SDK & Responses API: /resources/comparisons/self-hosted-ai-vs-openai-agents-sdk
- ibl.ai vs Vercel AI SDK: toolkit or owned platform: /resources/comparisons/self-hosted-ai-vs-vercel-ai-sdk
- ibl.ai vs Flowise: options after the 2026 sunset: /resources/comparisons/self-hosted-ai-vs-flowise
- ibl.ai vs Botpress: managed cloud or owned stack: /resources/comparisons/self-hosted-ai-vs-botpress
- Self-hosted AI vs cloud AI hosting: who runs what: /resources/comparisons/self-hosted-ai-vs-cloud-ai-hosting
- ibl.ai vs Dify: app builder or AI platform: /resources/comparisons/self-hosted-ai-vs-dify
- Self-hosted AI vs AWS Bedrock & Amazon Q: /resources/comparisons/self-hosted-ai-vs-aws-bedrock
- Self-hosted AI vs Azure OpenAI Service: /resources/comparisons/self-hosted-ai-vs-azure-openai
- ibl.ai vs OpenAI forward-deployed engineers: /resources/comparisons/ibl-ai-vs-openai-forward-deployed-engineers
- Autonomous agents vs RAG assistants & GPTs: /resources/comparisons/autonomous-agents-vs-rag-assistants
- OpenClaw as a LangChain alternative: /resources/alternatives/langchain-alternative
- OpenClaw as an OpenAI GPTs alternative: /resources/alternatives/openai-gpts-alternative
- OpenClaw as an AutoGPT alternative: /resources/alternatives/autogpt-alternative
- ibl.ai as a Cohere enterprise alternative: /resources/alternatives/cohere-enterprise-alternative
- ibl.ai as an AWS Bedrock alternative: /resources/alternatives/aws-bedrock-alternative
- ibl.ai as an IBM watsonx alternative: /resources/alternatives/ibm-watsonx-alternative

## What changed in ibl.ai recently?

Shipping velocity.

The platform ships a changelog roughly every week, grouped by service image. The three most recent:

- A Classifier Step That Routes Each Request to the Right Agent (2026-11-16): https://ibl.ai/updates/workflow-classifier-routes-each-request-to-the-right-agent — Agent workflows on ibl.ai gain a Classifier step: a decision model sorts each message into 2 to 20 categories, a yes or no, or a scale, sends it down that path, takes a Fallback below a confidence threshold (0.6 by default), and can be tried on a sample message before you publish.
- Platform Update — November 13, 2026 (2026-11-13): https://ibl.ai/updates/platform-update-2026-11-13 — Eleven services and packages moved in one week, led by a workflow classifier node that routes runs by model decision, an agent-chat MCP endpoint that runs a full agent turn as a Claude Desktop or claude.ai connector, agent channel identity linking with a Microsoft Teams connector, a credits-only billing flow with a plan picker, RBAC on the Create Agent form, a privacy audit log with PII and PHI filtering, and cross-organization agent copy.
- User Spend Limits: One AI Budget per Person, Every Agent (2026-11-10): https://ibl.ai/updates/user-spend-limits-default-for-new-members — A fourth Billing tab, User Limits, caps what one person spends across every agent in a workspace, and a Default for New Users gives each member who joins later their own limit automatically, with block or alert-only enforcement and near-limit alerts.

- All platform updates: /updates
- RSS: /updates/feed.xml

## What do I actually own when I build on ibl.ai?

A store of value, not a subscription.

Rented AI is a cost that recurs forever and leaves nothing behind. An owned platform is an asset: the code, the data, and the infrastructure are yours, and they keep their value after the contract ends.

### Rented AI

- Per-seat or per-token billing that scales with headcount
- Locked to one vendor's models and roadmap
- Your data lives in someone else's cloud
- Credits evaporate when the subscription lapses

### ibl.ai — owned

- Full source under a perpetual license
- Model-agnostic — run any LLM, switch anytime
- Code, data, and infra stay in your environment
- Usage-based, never per-seat — yours to keep

1.6M+ users across 400+ organizations run the platform this way, including NVIDIA, MIT, and Syracuse University. ibl.ai is family-owned and operated from New York, NY. Explore the self-hosted platform (/self-hosted-ai) or the full-code license (/full-code-license).

## Clone a repo and run it today

On ibl.ai you own all the code and the data, run it model-agnostic across any LLM, and pay with no per-seat pricing — so you can deploy anywhere, from your own cloud to a fully air-gapped network.

- github.com/iblai: https://github.com/iblai
- Book a technical demo: https://cal.com/iblai/30min

## Developer FAQ

### Is ibl.ai actually open source?

Yes — and we like to be precise about it. The application layer (iblai/os), the skills-intelligence app (iblai/lms), the developer toolkit, SDK, platform skills and MCP server (iblai/base), the infrastructure CLI (iblai/infra-cli), and the documentation itself (iblai/docs) are all public on GitHub.

Read them, fork them, and run them yourself — this page links each repository directly, and every documentation page under /developer is generated from the repository it documents.

There is one core platform component that isn't public open source. You can use it as a fully managed, hosted service, or take it under a perpetual source-code license and self-host the entire stack.

Either way the principle is the same: you own all the code and the data, and your deployment is yours — nothing is locked away from you.

Want to see exactly where every line is drawn? Talk to us and we'll walk you through the whole architecture.

### How do I start building on ibl.ai?

Install the skills with `npx skills add iblai/base --all` and type `/iblai-vibe-start` in Claude Code, Cursor or Codex: it asks what you are building, connects your organization, and scaffolds a production-ready Next.js app — SSO auth, agent chat, profiles, notifications and analytics wired to the platform.

Then add features one skill at a time from the catalog at /developer/skills, and deploy agents from the claws template library.

### Which ibl.ai skill should I start with?

`/iblai-vibe-start` for a new app, `/iblai-api-login` for a script or CI job that never signs a user in, and `/iblai-vibe-ops-init` to copy the starter.

After that, pick by the data your app touches: the six core families — custom data per user, the user's profile, memory, custom data per organization, agent configuration, and analytics — each have a skill with a screen (`iblai-vibe-*`) and a headless twin (`iblai-api-*`). The catalog at /developer/skills is searchable by what you are trying to do.

### What is the ibl.ai API, and where is the reference?

ibl.ai exposes a full REST API for the platform — agents, chat, users, analytics, documents, credentials, and more.

The interactive API reference lives at /developer/api-reference, the OpenAPI schema is served at /api/docs-schema, and the production API base is platform.iblai.app.

### What are agent skills and the MCP server?

The iblai/base repository ships agent skills plus a Model Context Protocol (MCP) server, so an AI agent can operate the entire ibl.ai platform through its REST API.

Install them with `npx skills add iblai/base --all` and your coding agent can create agents, manage users, and run workflows directly.

### Can I run the whole platform on my own infrastructure?

Yes. iblai/infra-cli is an interactive CLI that provisions the platform on your own AWS or GCP account with Terraform and configures servers with Ansible — and it can bootstrap any existing server.

You own all the code and the data, and the infrastructure too: cloud, VPC, on-premise, or fully air-gapped.

### Which LLMs can I use?

ibl.ai is model-agnostic. Run any LLM — Claude, GPT, Gemini, Llama, Mistral, Qwen, or your own fine-tune — and switch per workload without rebuilding your application.

You are never locked to a single vendor's models.

### How does this documentation stay accurate?

Every page under /developer/{repo} and /developer/guides is generated directly from the upstream GitHub repository by a sync script — the README and each file in the repo's docs directory.

That means there is no second copy to keep in sync: the engineers who ship the code write the documentation in the repo, and this site is the view. Pages that document the hosted platform backend, which has no public repository to mirror, live under /developer/platform and are maintained by hand.

### What does it cost, and what do I actually own?

Instead of per-seat SaaS, ibl.ai is usage-based or a flat perpetual license. What you build is a durable asset you own outright — not credits that evaporate when a subscription lapses.

ibl.ai serves 1.6M+ users from 400+ organizations and is family-owned and operated from New York, NY.
