Type: anthropic.agents.ClaudeAgent
Namespace: anthropic.agents
Description
Run Claude as an agent in a sandboxed environment with tool use capabilities. claude, agent, ai, anthropic, sandbox, assistant
Uses the Claude Agent SDK to run Claude with access to tools in a secure sandbox.
The agent can execute commands, read/write files, and use various tools while
maintaining security through sandbox isolation.
Use cases:
- Automated coding and debugging tasks
- File manipulation and analysis
- Complex multi-step workflows
- Research and data gathering
Properties
| Property | Type | Description | Default |
|---|---|---|---|
| prompt | str |
The task or question for the Claude agent to work on. | `` |
| model | language_model |
The Claude compatible model to use for the agent. | {"type":"language_model","provider":"empty","id... |
| system_prompt | str |
Optional system prompt to guide the agent’s behavior. | `` |
| max_turns | int |
Maximum number of turns the agent can take. | 20 |
| allowed_tools | list[str] |
List of tools the agent is allowed to use (e.g., ‘Read’, ‘Write’, ‘Bash’). | ["Read","Write","Bash"] |
| permission_mode | enum |
Permission mode for tool usage. | acceptEdits |
Outputs
| Output | Type | Description |
|---|---|---|
| text | str |
|
| chunk | chunk |
Related Nodes
Browse other nodes in the anthropic.agents namespace.