FastAPI for Agents

AgentOS turns your agents into a production API: 80+ endpoints for runs, sessions, memory, knowledge, and evals. Build your product on top of it, connect your agents to AI apps like Claude and ChatGPT, and chat interfaces like Slack, WhatsApp, Telegram, and Discord.

AgentOS gives you one AI backend for every frontend:

Your product. Call the REST API from your app: run agents, stream responses, and manage sessions, memory, and knowledge.AgentOS UI. Chat with agents, build new ones with AgentOS Studio, and inspect sessions, traces, memory, and evals from the AgentOS UI atos.agno.com.Coding agents. Manage the full agent development lifecycle (create, extend, improve, eval and review) using the skills in..agents/skills/

AI apps. MCP clients like Claude and ChatGPT can use your agents through the MCP server at/mcp

.Chat interfaces. Chat with your agents from Slack, WhatsApp, Telegram, and Discord.

Built on Agno. Everything runs in your cloud, your data lives in your database.

This codebase comes with:

Two platform agents that help you build and run the platform from your favorite AI apps like Claude and ChatGPT.Agent Builder creates agents, teams, and workflows using the AgentOS Studio.Platform Manager understands, monitors, and explains the platform: codebase questions, eval history, deployment checks, schedules.Coding-agent skills let Claude Code, Codex, Cursor, and other coding agents build, test, and improve the platform automatically — seeUsing the platform.

Trace data, agent code, evals, and system logs are all available to coding agents, so the platform can inspect and improve itself end to end.

The fastest way to get started is using a coding agent. Copy the prompt below into Claude Code, Cursor or Codex and it'll take you from zero to a running platform.

Help me set up AgentOS on this machine. Be self-driving: anything you can do — open a file, open a URL, launch an app — do it. Stop when you need a human: typing a secret, installing software, signing in to a website. When you do stop, tell me exactly what to do. Never read or print secrets.

1. Clone https://github.com/agno-agi/agentos-railway.git into a folder called agent-platform and cd in. Then read AGENTS.md end to end — it is the source of truth for how this platform works and answers most questions you'll hit along the way.
2. Confirm Docker is installed and running (`docker info` succeeds). If it's installed but not running, start it (`open -a Docker` on macOS) and poll until it's up. Stop for me only if Docker isn't installed — then give me steps to install Docker Desktop and wait.
3. Set up my environment. Run `cp example.env .env`, then help me set my OPENAI_API_KEY. If it's already set in my shell, tell me you found one and offer to copy it in — move the value across without reading or printing it. Otherwise open .env in my editor (cursor, code etc.) and ask me to paste the key. Never open a terminal editor like vim or nano from your own shell — it will hang this session.
4. Start the platform with `docker compose up -d --build`, then poll http://localhost:8000/docs until it returns 200 (first build takes a few minutes). If it never comes up, read `docker compose logs agentos-api` and fix what you find.
5. Prove it end to end with ./scripts/mcp_check.sh — it should print "MCP OK" and a real agent answer. Show me that answer: it's my platform talking.
6. Next, help me set up the AgentOS UI. Ask if you can open https://os.agno.com and walk me through the connection process: Connect OS → enter http://localhost:8000 → name it "Local AgentOS". That's where I chat with my agents and inspect sessions, memory, and evals.
7. Finish with a short summary and what I should do next: run /create-new-agent right here and you'll build a new agent with me, then /improve-agent to sharpen it — that loop is yours to keep. Mention in one line that I can also connect my platform to coding agents (like yourself) using `uvx agno connect`, and to claude.ai / ChatGPT over OAuth once it's deployed to production.

Prerequisite:[Docker]installed and running.

git clone https://github.com/agno-agi/agentos-railway.git agentos
cd agentos

cp example.env .env

docker compose up -d --build

Confirm your AgentOS is running at http://localhost:8000/docs.

• Open os.agno.comand sign in. - Click Connect OS, enterhttp://localhost:8000

as the URL, name itLocal AgentOS, and connect.

• Click Chat under theAgent Builder agent and try the first prompt: "Build an agent that tracks AI news and writes a daily brief". Go through the agent development process. - Once created, click the Refresh button on the top right. You should now see the "Daily AI News Brief" agent in theAgents dropdown. Click the newly created agent. - Ask: "What's new with Anthropic?"

Click Chat under Platform Manager and ask: "How healthy is the platform?" It answers from the codebase and runtime data — eval history, deployment checks, schedules, and the component you just built.

You can run the platform anywhere that supports containerized images. This codebase comes with scripts to deploy the platform to Railway.

Prerequisite:[Railway CLI]installed andrailway login

completed.

Create a new .env.production

file for production credentials.

cp .env .env.production          # or cp example.env .env.production

Keeping a separate .env.production

lets us use different values for local and production: different OpenAI keys, production-only credentials, a different Slack workspace.

./scripts/railway/up.sh

This provisions the AgentOS service and Postgres on the same private network. The script s and asks for a JWT verification key for authentication (see next section).

Token-Based Authorization is on by default. Without a JWT_VERIFICATION_KEY

or JWT_JWKS_FILE

, the app refuses to serve traffic in production. The platform's job is to keep your data private, so the safe default is "refuse to start" without an authentication token.

Token-Based Auth gives you three things:

No public access. The server rejects requests without a valid token.Per-request identity. Middleware parses the token and extracts theuser_id

,session_id

, and custom claims. Each request is tied to a user and session, giving you auditability and traceability.Granular permissions. User tokens can run an agent and view their own sessions. Admin tokens read everyone's sessions and test any agent.

During ./scripts/railway/up.sh

, the script creates your Railway domain and s so you can mint the key before the app starts.

• Open os.agno.com, clickConnect OS→** Live**, enter your Railway domain, and connect. - Name it Live AgentOS. - Go to Settings→** OS & Security**. - Turn Token-Based Authorization (JWT) on. - Copy the public key.
• Paste the full public key into the up.sh

prompt. The script saves it into your env file for future syncs:

JWT_VERIFICATION_KEY="-----BEGIN PUBLIC KEY-----
MIIBIjANBgkq...
-----END PUBLIC KEY-----"

Heads up.Live AgentOS Connections are a paid feature. UsePLATFORM30

to get 1 month off. We are working on a free trial so you don't have to pay to try.

If you get something wrong, you can re-sync environment variables with ./scripts/railway/env-sync.sh

.

Re-run uvx agno connect

, this time pointed at your deployed domain, to connect Claude Code, Claude Desktop, Codex, and Cursor to your production platform:

uvx agno connect --url https://<railway-domain>

For claude.ai and ChatGPT (web): add https://<railway-domain>/mcp

as a custom connector in the chat app's connector settings. Leave the form's optional OAuth fields (client ID / client secret) empty. Click Connect and, on the consent page, enter the MCP_CONNECT_SECRET

that up.sh

generated during deploy (saved in .env.production

).

You can check the logs on the Railway dashboard, or by running the following command:

railway logs --service agent-os

To redeploy your AgentOS, run the following command:

./scripts/railway/redeploy.sh

Recommended: Auto-deploy on merge to main

using:

• Open the Railway dashboard, your project, the agent-os service, Settings. - Under Source, click** Connect Repo**and pick your repo. - Set the deploy branch to main

and save.

Push to main

triggers a build and rolling deploy. ./scripts/railway/env-sync.sh

is still how you sync env changes.

To re-sync environment variables, run the following command:

./scripts/railway/env-sync.sh
./scripts/railway/down.sh

Deletes the Railway project: the agent-os service, the pgvector database, and its volume, including all data. It also removes the AGENTOS_URL from your env file so a future up.sh

starts clean.

Set authorization=False

in app/main.py and redeploy. Use this only inside a private VPC behind another auth layer. Without it, anyone who guesses your Railway domain can access your platform.

This platform is designed so that coding agents can drive the entire create → improve → evaluate → maintain lifecycle for you.

Open your coding agent of choice (Claude Code, Codex, Cursor) and run:

/create-new-agent

It asks a few questions, generates the agent file in agents/

, registers it in app/main.py

, adds its description and quick prompts to app/config.yaml

, restarts the container, and smoke-tests it live.

Improve your agents by running the following skills:

— Add a tool, add a capability, refine the instructions, fix a known bug./extend-agent

— Claude simulates scenarios from the agent's/improve-agent

INSTRUCTIONS

, runs them against the live container, judges the responses, and edits until they pass.

Run the eval suite to check for regressions. The evals live in evals/cases.py, and run history shows up at os.agno.com next to your sessions and traces.

The evals run on the host machine, so set up the venv with ./scripts/venv_setup.sh && source .venv/bin/activate

, then:

python -m evals --tag smoke      # fast checks of the self-driving surfaces
python -m evals --tag release    # broader pre-release confidence
python -m evals --name <case>    # one case while iterating
python -m evals -v               # stream the full run with rich panels

If a case fails, run ** /eval-and-improve** — it diagnoses each failure, fixes what's in scope, and loops until green.

Because the repo is managed by coding agents, it moves fast. Run /review-and-improve

before a release or after a refactor: it sweeps for drift between docs, code, and config, auto-fixes mechanical drift like stale paths and missing env vars, and flags anything bigger.

AgentOS comes with an MCP server at /mcp

(enabled by setting mcp_server=True

in app/main.py), so any MCP client can call your agents, teams, and workflows through tools like

run_agent

, run_team

, and run_workflow

.Register your AgentOS with the MCP clients on your machine:

uvx agno connect

It auto-detects Claude Code, Claude Desktop, Codex, and Cursor and registers http://localhost:8000/mcp

. After a successful connection, open one of these apps and ask:

can you access my agentos mcp?

claude.ai and ChatGPT (web). Hosted AI apps reach your platform over the internet and need an OAuth login. Deploy to production (above), add https://<domain>/mcp

as a remote connector, and approve the consent page with your connect secret.

Variable	Required	Default	Description
OPENAI_API_KEY
yes	none	OpenAI key for models and embeddings.
RUNTIME_ENV
no	prd
dev disables JWT. Compose sets this to dev for local — never put it in an env file that syncs to Railway, or production deploys unauthenticated.
JWT_VERIFICATION_KEY
prd	none	Public key from os.agno.com. Required when RUNTIME_ENV=prd , unless JWT_JWKS_FILE is set.
JWT_JWKS_FILE
prd	none	Path to a JWKS file; alternative to JWT_VERIFICATION_KEY for production JWT verification.
AGENTOS_URL
no	http://127.0.0.1:8000
Scheduler base URL. scripts/railway/up.sh auto-sets it to your Railway domain; set by hand only for a custom domain or tunnel. Also the public origin OAuth metadata derives from when MCP_CONNECT_SECRET is set.
MCP_CONNECT_SECRET
no	none	If set (≥16 chars, e.g. openssl rand -base64 32 ), /mcp becomes its own OAuth 2.1 authorization server so claude.ai and ChatGPT (web) can connect; connecting asks for this secret on a consent page. Requires AGENTOS_URL . scripts/railway/up.sh auto-generates it on deploy. PAT and JWT bearers keep working alongside.
AGENTOS_MCP_SIGNING_KEY
no	none	Optional high-entropy signing-key material (≥32 chars) for OAuth tokens. Unset, a strong key is generated and persisted in the database. Rotating it invalidates outstanding tokens.
ENABLE_DEPLOY_CHECK
no	True
The reference deployment-check cron runs daily by default. Set False to disable; the workflow is runnable on demand regardless.
ENABLE_SCHEDULED_EVALS
no	False
If True , schedules the run-evals workflow daily. Off by default because it uses model calls.
EVALS_TAG
no	smoke
Eval tag run by the run-evals workflow.
EVALS_CASE_TIMEOUT_SECONDS
no	90
Default per-case timeout for run-evals runs; applies only to cases that don't set their own timeout_seconds .
EVALS_SUITE_TIMEOUT_SECONDS
no	900
Whole-suite timeout for run-evals runs; per-case timeouts are the granular limit. The default bounds the smoke tag's worst case (incl. builder-case teardown).
PARALLEL_API_KEY
no	none	Authenticates the WebSearch Agent's Parallel SDK / MCP connection.
SLACK_BOT_TOKEN / SLACK_SIGNING_SECRET
no	none	Both must be set to enable the Slack interface.
DB_HOST / DB_PORT / DB_USER / DB_PASS / DB_DATABASE
no	matches compose	Postgres connection.
DB_DRIVER
no	postgresql+psycopg
SQLAlchemy driver.
AGNO_DEBUG
no	False
If True , Agno emits verbose debug logs. Compose sets this for dev.
WAIT_FOR_DB
no	False
If True , the entrypoint blocks on the DB before starting. Compose sets this.

Agno documentationAgentOS introductionAgno on GitHub. Drop a star if this is useful.

source & further reading

github.com — original article