> Source: https://wexa.ai/docs/agents/overview

# Agents

An agent is one LLM worker: a single set of instructions, tools and a model. A peer agent is an external agent on another platform that a process flow can hand a step to. A model is the LLM an agent runs on. An execution is one run, with its inputs, outputs and trace. You create an agent, run it with a goal, and poll the resulting execution for the outcome. Creating or changing an agent needs `fabric:orchestrate.write`, reading one needs `fabric:orchestrate.read`, and running one needs `fabric:agent.run`.

Start with `create-agent`, which returns the `agent_id` the others take.

- [create-agent](/docs/agents/create-agent): Create an agent.
- [get-agent](/docs/agents/get-agent): Read one agent.
- [list-agents](/docs/agents/list-agents): List the project's agents.
- [update-agent](/docs/agents/update-agent): Change an agent's fields.
- [delete-agent](/docs/agents/delete-agent): Delete an agent.
- [run-agent](/docs/agents/run-agent): Run an agent with a goal.
- [agent-versions](/docs/agents/agent-versions): Read promoted version history.
- [rollback-agent-version](/docs/agents/rollback-agent-version): Make an earlier version live.
- [register-peer-agent](/docs/agents/register-peer-agent): Register an external agent.
- [handshake-peer-agent](/docs/agents/handshake-peer-agent): Test an external agent endpoint.
- [list-peer-agents](/docs/agents/list-peer-agents): List peer agent connections.
- [delete-peer-agent](/docs/agents/delete-peer-agent): Delete a connection.
- [list-models](/docs/agents/list-models): List models and the default.
- [run-model](/docs/agents/run-model): Call a model directly.
- [set-model](/docs/agents/set-model): Set the organization default.
- [get-execution](/docs/agents/get-execution): Read one run.
- [list-executions](/docs/agents/list-executions): List runs.
- [pause-execution](/docs/agents/pause-execution): Pause a run.
- [resume-execution](/docs/agents/resume-execution): Resume a paused run.
- [restart-execution](/docs/agents/restart-execution): Re-run an execution.