Build domain-specific long-horizon agents.
Cayu is an open-source Python framework for building and running domain-specific long-horizon agents, with a durable runtime built in. Compose your harness around the models, tools, knowledge, and business rules your job needs.
pip install cayuBuild the agent. Shape how it works.
Cayu provides the Python components and defaults for your harness, with an integrated runtime to execute it. You supply the domain knowledge and define how the agent does its job.
| Concept | Its role in Cayu |
|---|---|
| Framework | The Python structure, APIs, defaults, and components you build with. |
| Harness | The system you assemble with Cayu: context, tools, memory, policies, domain logic, and verification. |
| Runtime | The integrated execution machinery: sessions, state, workers, recovery, approvals, and budgets. |
| Agent | The resulting product that performs work in your domain. |
Build with the framework. Assemble your harness. Run on the integrated runtime. The result is your agent.
A clear goal. Hours of work to get there.
“Process these 5,000 documents” has a finish line. Getting there may involve thousands of calls, new evidence, a human decision, and a restart.
This example shows the lifecycle you can build with Cayu. Your harness defines the business rules and completion checks.
Python you can make your own.
Start with a project that has a place for your prompts, tools, policies, memory, and evals. Bring the libraries and domain code your team already uses.
Python 3.11+ and uv · Run locally or on your infrastructure
uvx cayu new support-agent
cd support-agent
uv sync --extra devuv run --no-sync cayu check --fail-on warning --json
uv run --no-sync pytest
uv run --no-sync cayu eval runConfigure your model provider in configuration/settings.py or CAYU_PROVIDER first.
uv run --no-sync python run.py --message "Review this change."A scaffold with tests and an evaluation to extend.
Open the generated project in your development tool, or start with this prompt and build and refine your agent together.
The foundation behind the work.
Build with your models and tools.
Choose providers, connect Python and MCP tools, use web and browser capabilities, and compose context, knowledge, and memory.
Coordinate and carry the work through.
Delegate across agents and workspaces. The integrated runtime records progress, enforces permissions and budgets, and supports approvals and recovery.
Understand and improve.
Inspect the work, trace model costs, and evaluate behavior using the path the agent took and the result it produced.
Your agent. Your way to run it.
Build your agent and its domain-specific harness with Cayu. Run it on your own infrastructure, or use Cloud as your team’s shared workspace.
Built from the work.
Cayu was extracted from the production runtime behind an agent-operated software factory that built and deployed thousands of business applications. Specialized agents worked together as its SRE, product manager, coder, and forward-deployed engineering assistant.
Agent SDKs and frameworks made the model-and-tool loop quick to build. Production quality needed control of that loop and of everything around it: durable state, credentials, execution environments, human intervention, recovery, cost attribution, and evaluation. Cayu brings those together.
Meet CayuBring your domain. Build your agent.
What does your agent need to do? Join the conversation