# AgentDeck SDK

> The production runtime for agents you already have — composition, durable human-in-the-loop approvals, sessions, streaming, run control and one ordered event log, wrapping the OpenAI Agents SDK and LangGraph.

AgentDeck SDK is a Python harness for agents you have to operate. You write agents, workflows and
skills as small declarations in a `.agentdeck/` directory; AgentDeck supplies everything around
them — discovery, layered settings, sessions, streaming, MCP servers, durable human-in-the-loop
approvals, run control, and one ordered event log per run.

It wraps the OpenAI Agents SDK and LangGraph rather than replacing them: an `Agent` compiles to an
SDK agent, a `Workflow` compiles to a LangGraph graph, and each is executed by its own engine.
AgentDeck owns configuration; the engines own execution. There is no agent loop here and no graph
engine of its own.

Use it when the wiring around an agent has become the work: several agents and workflows in one
project, a chat surface and a batch path over the same definitions, runs you must inspect
afterwards, approvals that outlive the process that requested them.

## Documentation

- [AgentDeck](https://agentdecksdk.com/): Write the agent. AgentDeck owns everything around it.
- [Getting Started](https://agentdecksdk.com/getting-started): Install AgentDeck, point it at a model, and run your first agent.
- [Core Concepts](https://agentdecksdk.com/concepts): The things you write, and what AgentDeck does around them.
- [Agents](https://agentdecksdk.com/concepts/agents): Declarative agent definitions — keyword arguments in, SDK Agent out.
- [Capabilities](https://agentdecksdk.com/concepts/capabilities): Sandbox powers — not part of v3, and no longer in the codebase.
- [Choosing a Store Backend](https://agentdecksdk.com/concepts/choosing-a-store-backend): Four independent storage decisions, one environment variable each, and what each default costs you.
- [Protocols and Surfaces](https://agentdecksdk.com/concepts/protocols-and-surfaces): What reads the canonical event stream today, and, plainly, what does not exist yet.
- [Run Control](https://agentdecksdk.com/concepts/run-control): What a safe point is, why asking a run to stop is not the same fact as it stopping, and what a resume replays.
- [Runs and the Event Log](https://agentdecksdk.com/concepts/runs-and-the-event-log): The one ordered record every turn appends to, and why a run's status is read from it instead of stored anywhere else.
- [Sessions and Memory](https://agentdecksdk.com/concepts/sessions-and-memory): What a session is, where its history lives, and why it is not the event log.
- [Skills](https://agentdecksdk.com/concepts/skills): Progressive knowledge disclosure into an agent's own execution, described by SKILL.md.
- [Workflows](https://agentdecksdk.com/concepts/workflows): Typed state, a LangGraph graph, and durable pauses for humans.
- [Add a Tool](https://agentdecksdk.com/guides/add-a-tool): Give an agent a plain Python function it can call.
- [Human Approval](https://agentdecksdk.com/guides/human-approval): Pause a workflow for a person's decision, and answer it from wherever that person is.
- [Serve Over HTTP](https://agentdecksdk.com/guides/serve-over-http): Run agentdeck behind FastAPI, and talk to an agent from outside the process.
- [Pause, Resume, Cancel](https://agentdecksdk.com/operating/pause-resume-cancel): Which endpoint to call to stop a run you did not start, what each response means, and how to tell whether it worked.
- [Reference](https://agentdecksdk.com/reference): Technical reference for AgentDeck's public APIs and configuration.
- [Capabilities](https://agentdecksdk.com/reference/capabilities): The sandbox capability specs — removed; sandboxing is not part of v3.
- [CLI](https://agentdecksdk.com/reference/cli): The agentdeck command tree, generated from agentdeck/cli.py's own --help output.
- [Deck](https://agentdecksdk.com/reference/deck): The composition root — builds a catalog from Agent/Workflow objects or a project directory, and runs every turn on it.
- [Definitions](https://agentdecksdk.com/reference/definitions): Agent, AgentDeclaration, Workflow, and WorkflowDeclaration — the authoring API, field by field.
- [Settings](https://agentdecksdk.com/reference/settings): Every AGENTDECK_* (and OPENAI_*/TAVILY_*) environment variable, generated from agentdeck/runtime/settings.py.
- [Roadmap](https://agentdecksdk.com/roadmap): What AgentDeck is working on next, in the order it is being worked on, and what is deliberately not planned.
- [Known Issues](https://agentdecksdk.com/known-issues): Defects in the current release that will surprise you, what happens, and what to do until each is fixed.
- [Changelog](https://agentdecksdk.com/changelog): What changed in each release of AgentDeck, and what to do about it when upgrading.

## Source

- [Repository](https://github.com/agentdecksdk/agentdeck): issues, releases and examples
- [Full documentation as one file](https://agentdecksdk.com/llms-full.txt)
