> ## Documentation Index
> Fetch the complete documentation index at: https://docs.groundforge.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Agents

> Create a managed Agent or inspect an external Agent connected through the SDK.

Open **Agents** in the selected Workspace.

## Managed Agent

Choose **Create Agent**, select the managed type, and enter the model, write-only model credential, system instructions, Service Account, Channels, and Gateways. Save and publish the Agent after attaching a Channel.

The model credential is stored as an Agent-owned credential and is not shown again in full.

### Example

Use [Weather MCP Example](/examples/weather-mcp) as the capability source. Create a managed weather Agent, attach the Weather MCP Gateway, publish it, and ask for a forecast through a Channel.

## External Agent

Create the external Agent identity first, then initialize the SDK in your running application. The Agent page shows connection state, Discovery, Channels, Gateways, Traces, approvals, and Analytics.

### Example

<Card title="groundforgeai/demo" icon="github" iconType="brands" href="https://github.com/groundforgeai/demo" horizontal cta="Open repository">
  Connect `simple_agent_service.py` and confirm `get_weather` appears in Discovery.
</Card>

> **Screenshot placeholder**\
> Add `/images/platform/agent-overview.png` showing Agent identity, connection or model state, Channels, Gateways, Discovery, and Analytics.

See [Agents concept](/concepts/agents) and [LangChain Agent Example](/examples/langchain-agent).

## Example paths

Use [Weather MCP Example](/examples/weather-mcp) with a managed Agent or [LangChain Agent Example](/examples/langchain-agent) with an external Agent.


## Related topics

- [Agents](/concepts/agents.md)
- [Get Agent](/api-reference/endpoints/agents/get-agent.md)
- [Publish Agent](/api-reference/endpoints/agents/publish-agent.md)
- [Update Agent](/api-reference/endpoints/agents/update-agent.md)