# Astropods > Every agent your company runs, in one operating system. Deploy from a single spec, trace every run without instrumenting it, and set what each agent can reach and spend. One platform to govern every agent. Astropods brings every agent into a single operating system: governed centrally, run everywhere, and measured against the work your team already does. ## Where organizations are Every organization is somewhere on the path to AI-native. ### A flock of chaos Your employees are building their own agents, installing Slack bots, and using all the latest AI models. Your token budgets are stressed and it’s impossible to track where and how your company’s data is being used. ### Getting in formation Your agents are unified on a single platform with access control, budgets, and shared memory and context. Every token is accounted for and new agents can be deployed with confidence. ### A global fleet Your agents scale automatically, backed by industry standard security and performance. Cost optimizations and governance policies make it effortless and safe to deploy agent fleets of any size. ## Anything you ship on Astropods is a governed agent. Astropods is the layer between your code and production. It gives you the spec, the guardrails, and the record of every run. You bring the work you want done. - **One view for all your agents** — Whether your team builds an agent or brings one in, it lives in one record. Conversations distill into patterns, so you can see what people actually ask for. - **Production-grade controls** — Decide at deploy what an agent can reach, what needs a human to sign off, how much it can spend, and where its data stays. It runs under the same rules as the rest of your production software. - **Access set per user or team** — You decide who can build an agent, who can deploy one, and who can use it. Permissions follow the roles and teams you already have, so you are not wiring up access agent by agent. ## Manage your intelligence budget from a central place. See human and agent work in one view. Where agents add the most value, and where your teams get the most leverage from them. ## Set up automatic rules. Astropods ensures every agent follows them. Choose what to watch for, which agents it applies to, and what should happen. An agent that starts to spend too much is stopped instead of showing up on the bill. ## Trust is built right into the foundation. Agents reach your most sensitive systems. Astropods gives them the same controls you already apply to production software, and the record to prove it. - **Your data is yours** — Prompts, traces, and evals stay in your cloud, under your contracts. - **Least privilege by default** — Each agent gets only the scopes its task needs, enforced at the platform layer. - **Every run leaves a record** — Who asked, what the agent did, and what it cost, in order. - **No version reaches your users unproven** — Each release is measured against the one live today. If quality drops, the release is held, or rolled back on its own. - **Run inside your boundaries** — Managed, your own VPC, or fully isolated for regulated work. Wherever you prefer. ## Pricing Pricing that grows with your agent team. Start for free and pay as-you-go or talk to us about a custom plan that works for you. ### Hobby — Free to start For the solo builder or small teams shipping their first agents. Free signup credits to start. No card needed until you use them. - Create and run agents, and pay for what you use - Invite your team and build faster together - Track every dollar each agent spends - Private knowledge stores with AWS PrivateLink - Download agent training datasets - Community and email support ### Enterprise — Custom For organizations running many agents across many teams. We build the plan around your team and your goals. - Astropods engineers help you build and onboard - Create and run fleets of AI agents - Dedicated clusters for your agents and data - Single sign-on (SSO) - Per-team policies - Dedicated support and SLA ## Postman Agent Stack - [Passport](https://app.usepassport.ai/) - [Fabric](http://app.fabricgateway.ai) - [Astropods](https://astropods.com/) - [Fern](https://buildwithfern.com/) - [Postman](https://www.postman.com) # Astropods for builders Define your agent. We do the rest. Install the CLI, scaffold an agent, and push it to the registry. Astropods handles containers, secrets, scaling, and observability from the first run. Install: `curl -fsSL https://astropods.com/install | sh` ## The spec is the deploy form. Declare an input, a model, or a provider in `astropods.yml` and it becomes a field the moment the blueprint deploys. ## Bring the framework you already use. An adapter wraps your agent where it stands. The code keeps its own shape, the spec adds the runtime around it, and nothing gets rebuilt to ship. ## Build the agent. The rest is already running. No registry to build, no tracing to bolt on, and no eval harness to write. These are built in for every blueprint. - **Deploy multiple instances of one agent** — Push an agent blueprint once and the whole team can find it, run it, and build on it. Each deploy is one click and its own agent, with its own settings and its own spend. - **No instrumentation required** — Every run is traced automatically. That includes each step, each tool call, token usage, and cost. No SDK to add, no exporter to configure. - **Continual regression testing** — Write the checks you care about in an eval spec and a judge model runs them on real conversations. Keep the runs that matter as a dataset, and score the next version against it before it ships. ## All the infrastructure you need, already there. No store to provision, no model key to rotate, no token in your repo. The blueprint arrives and the runtime is already there. ## Explore agent blueprints Use a blueprint that already works, add your own keys, and deploy it under your account. # Astropods for startups Your first agent. And your hundredth. Startups get the full Astropods platform, plus free credits and support from our engineers to help you get started. ## Why startups choose Astropods. Whether you are shipping your first agent or your fiftieth, Astropods gets you up and running. - **Up to $10,000 in credits** — We set you up with credits to get started building and running agents with your team. - **Support from our engineers** — A shared channel with Astropods engineers, and help to get the first agent from spec to production. - **Production controls on day one** — Single sign-on, audit logs, and a spend limit per agent. The same controls the enterprise plan has. ### Who can apply - Less than $5M raised - Incorporated in the last three years - Fewer than 25 people, not a customer today Not sure you qualify? Apply anyway and tell us about the team. ## Questions you might have ### What makes Astropods different from other agent platforms? Most agent tools are point solutions built for developers. Astropods is an operating system: every agent lives in one governed environment, shares context and memory, and is measured against the work your team already does. ### Do I own my data? Yes. Your proprietary data, prompts, traces, and evals stay in your cloud under your contracts. They are never used to train external models and can be exported or deleted on your terms. ### Which models, frameworks, and clouds can I use? Any major model provider and framework you already use. Astropods runs on the cloud you choose, including your own VPC, with no lock-in. ### How do I measure agent ROI? Astropods tracks cost, usage, and outcomes in one view. You can see where agents add the most value, where your teams get leverage, and where spend is going to waste. ### Can I bring in external coding agents and track them alongside everything else? Yes. Connect the coding agents your team already uses — Claude Code and Codex today — so their output, cost, and usage flow into the same observability layer. Every agent, internal or external, is tracked in one place. ### Can I run agents in my own cloud? Yes. Enterprise supports bringing your own cluster and running inside your VPC, for security and data sovereignty. ### What if I don’t have any agents running yet? Most of our customers started exactly there. Astropods helps you identify the highest-value first use case, build or connect your first agent, and prove ROI before you scale. You don’t need a fleet of agents to get value. You need one agent running with the right governance and measurement. ### Is this secure enough for regulated industries? Yes. Least-privilege access, audit logging, eval-driven safety controls, and automatic rollbacks are built in, so you can deploy agents with the same rigor as any production workload. ## Questions builders ask ### What do I need before the first deploy? An account and Docker. Install the CLI, run `ast project create`, and set the credentials it asks for. Astropods builds the image and runs it. There is no cluster to stand up first. ### What is a blueprint? One file, `astropods.yml`, describes the whole agent: where it builds from, the models it calls, what it can reach, and what somebody has to fill in. Push it and you have a blueprint, a versioned snapshot in the registry. Deploy that and you have a live agent, which anyone in the organization can run or start from without opening the YAML. ### Can I keep the framework I already use? Yes. There are adapters for Google ADK, OpenAI, CrewAI, LangChain, LangGraph, the Vercel AI SDK, Mastra, and more. If you already have a Dockerfile, point the spec at it and Astropods takes the runtime around your code. ### What if I already have agents in production? Bring them over. There is no rewrite: the agent keeps its code, its framework, and the Dockerfile you already ship. One file beside it is the whole port, and it can run next to what you have today until you are ready to move the traffic. ### Can I run the agent locally first? Yes. `ast project start` runs the agent on your machine against the same spec the deploy reads, so what you test is what you push. ### Can I serve my own frontend for my agent? Yes, and it is as easy as adding one line to the spec. Set `frontend: true` and the agent gets its own HTTPS hostname, with sign-in already at the front door. Anything that speaks HTTP will do. ### Where do secrets live? In the account vault. The spec refers to them by name, as `KEY=@SECRET_NAME`, so no value is in your repo or your image. The agent gets them at run time. ### Can I run this in my own cloud? Yes. Astropods runs managed, in your VPC, or fully isolated for regulated environments. The spec and the CLI do not change. ## Questions startups ask ### What do the credits cover? Platform usage — the agents you deploy, the runs they make, and the traces and evals they produce. Model tokens you buy from a provider are billed by that provider. ### How long does the review take? We try to get back within 2 business days. We read every application, and we write back either way. ### What happens when the credits run out? You pay for the usage you add from then on. ### Can I keep the framework I already use? Yes. There are adapters for Google ADK, OpenAI, CrewAI, LangChain, LangGraph, the Vercel AI SDK, Mastra, and more. If you already have a Dockerfile, point the spec at it. ### We have no agents running yet. Is it too early? No. Most teams in the program start there. One agent in production, measured, is the point. ### Do I own my data? Yes. Your prompts, traces, and evals stay yours, they are never used to train external models, and you can export or delete them on your terms.