| name | connecting-agents |
| description | Delegates tasks to Manus when local capabilities are insufficient. Use when tasks require real browser automation, web scraping, persistent compute, long-running autonomous execution, or multi-step workflows with deliverables. |
| allowed-tools | WebFetch, Bash, Read, Write |
Connecting Agents
Extend capabilities by delegating to external agents.
When to Delegate
Delegate when:
- Task requires capabilities Claude lacks (real browser, persistent compute, OS access)
- External agent has specialized domain expertise
- Long-running autonomous execution needed
- User explicitly requests external agent
Handle locally when:
- Claude can complete the task directly
- Latency/cost of delegation exceeds benefit
- Task requires conversational back-and-forth
Supported Agents
Manus
Complete AI agent with autonomous multi-step execution.
Capabilities: Browser automation, file system, research, content creation, data processing
Use for: Web scraping, form filling, complex research, deliverables (reports, presentations, websites)
API: RESTful. Docs
import requests
response = requests.post(
"https://api.manus.im/v1/tasks",
headers={"Authorization": f"Bearer {MANUS_API_KEY}"},
json={"task": "Research competitor pricing and create summary report"}
)
task_id = response.json()["task_id"]
Delegation Process
- Identify capability gap
- Select appropriate external agent
- Formulate task description
- Execute via API/MCP
- Monitor for completion
- Integrate results into current context
- Report outcome to user
Integration
Results from external agents should be:
- Summarized concisely
- Validated if possible
- Presented with source attribution
- Flagged if uncertain or incomplete
