Common tasks
Find the task that matches your next step. Follow one path until you have a working first version, then return to the reference pages for detail.
Build an agent#
| Task | Start here |
|---|---|
| Learn the language | Language basics |
| Call a model | LLM calls and agents |
| Give an agent tools | LLM tools |
| Add several dependent stages | Workflow runtime |
| Run independent child agents | Delegated workers |
| Test without an API key | Mock LLM responses |
Connect a system#
| Task | Start here |
|---|---|
| Call an MCP server | MCP, ACP, and A2A integration |
| Expose a Harn program | Outbound workflow server |
| Receive events | Trigger manifests |
| Author a connector package | Connector authoring |
| Add human approval | Human in the loop |
Operate a run#
| Task | Start here |
|---|---|
| Inspect a run | Harn portal |
| Debug a failed agent | Debugging agent runs |
| Replay or evaluate behavior | Testing |
| Deploy an orchestrator | Orchestrator |
| Manage secrets and OAuth | Orchestrator secrets |
Starting from zero#
- Follow Getting started.
- Read Language basics.
- Build one small model-backed program with LLM calls and agents.
- Add tools, workflows, or workers only when the task needs them.
- Read Best practices before you run unattended or write to an external system.