| name | manage-agents |
| description | List, run, and monitor ohwow AI agents. Use when the user wants to execute an agent, check task results, or see what agents are available. |
| tools | ["mcp__ohwow_ohwow_list_agents","mcp__ohwow_ohwow_run_agent","mcp__ohwow_ohwow_get_task","mcp__ohwow_ohwow_list_tasks","mcp__ohwow_ohwow_workspace_status"] |
Agent Management
You are helping the user manage their ohwow AI agents. Follow this workflow:
Step 1: Discover
If the user hasn't specified an agent, list available agents first:
- Use
ohwow_list_agents to see all agents with their roles and capabilities
- Present them in a concise table: name, role, status
Step 2: Execute
When the user picks an agent or describes a task:
- Use
ohwow_run_agent with the agent ID and a clear prompt
- Note the returned task ID
Step 3: Monitor
After running an agent:
- Use
ohwow_get_task to check if the task completed
- If still running, let the user know and offer to check again
- When complete, summarize the result
Error Recovery
- If
ohwow_run_agent returns an error, check process status with ohwow_workspace_status
- If the task shows status "failed", read the error details in the
ohwow_get_task output
- For long-running agents, poll
ohwow_get_task every 10-15 seconds until status is "completed" or "failed"
Tips
- If the user says "run" or "execute" without specifying an agent, use
ohwow_list_agents first to help them pick
- For complex multi-step work, suggest running multiple agents with
ohwow_chat which has access to spawn_agents and await_agent_results
- Use
ohwow_workspace_status if the user asks about overall system health
$ARGUMENTS