Shayan Rais
127 lines
5.1 KiB
Markdown
127 lines
5.1 KiB
Markdown
# Agent Orchestration Best Practices
|
|
|
|
## Problem: Sub-agents Not Invoking
|
|
|
|
### Issue Description
|
|
|
|
When creating orchestrator agents that coordinate multiple sub-agents, a common mistake is using bash commands or other tools instead of the proper `Task` tool to invoke sub-agents. This results in the sub-agents not being invoked at all.
|
|
|
|
### Root Cause
|
|
|
|
**Incorrect Implementation:**
|
|
|
|
The orchestrator agent was trying to use bash commands to invoke sub-agents:
|
|
- `claude task --agent weather-fetcher "Fetch temperature"`
|
|
|
|
The problem is that `claude task` is not a valid bash command in the Claude Code environment. Agents cannot invoke other agents through bash/CLI commands. Instead, they must use the `Task` tool programmatically.
|
|
|
|
### Solution
|
|
|
|
**Correct Implementation:**
|
|
|
|
1. **Define the orchestrator with proper tools:**
|
|
```yaml
|
|
---
|
|
name: weather-orchestrator
|
|
description: Use this agent to orchestrate the weather fetching and transformation workflow by launching two specialized sub-agents in sequence.
|
|
tools: Task
|
|
model: haiku
|
|
color: green
|
|
---
|
|
```
|
|
|
|
2. **Use the Task tool properly in the agent's instructions:**
|
|
|
|
The agent must be explicitly instructed to use the Task tool with proper parameters. Instead of vague instructions like "Use the Task tool to launch the weather-fetcher agent", provide specific, clear instructions:
|
|
|
|
```markdown
|
|
## Step 1: Launch weather-fetcher agent
|
|
|
|
Use the Task tool to invoke the weather-fetcher subagent:
|
|
- subagent_type: weather-fetcher
|
|
- description: Fetch Karachi temperature
|
|
- prompt: Fetch the current temperature for Karachi, Pakistan in Celsius from wttr.in API. Return the numeric temperature value in your final report.
|
|
- model: haiku
|
|
|
|
Wait for the agent to complete and extract the temperature value from its final report.
|
|
```
|
|
|
|
3. **Key Requirements for Orchestrator Agents:**
|
|
|
|
a. **Explicit Tool Usage**: State clearly "DO NOT use bash commands or any other tools. You must use the Task tool to invoke sub-agents."
|
|
|
|
b. **Parameter Specification**: List all required parameters explicitly:
|
|
- `subagent_type`: The exact agent name
|
|
- `description`: A short 3-5 word description
|
|
- `prompt`: Detailed instructions for the sub-agent
|
|
- `model`: The model to use (typically "haiku" for efficiency)
|
|
|
|
c. **Sequential Execution**: For sequential workflows, explicitly state "Launch agents one at a time, wait for completion before launching the next."
|
|
|
|
d. **Data Passing**: Provide clear instructions on how to extract data from one agent's report and pass it to the next agent's prompt.
|
|
|
|
### Before and After Comparison
|
|
|
|
#### Before (Broken):
|
|
```markdown
|
|
## Your Task
|
|
|
|
1. **Launch weather-fetcher agent**: Use the Task tool to launch the weather-fetcher agent
|
|
- This agent will fetch the current temperature for Karachi, Pakistan in Celsius
|
|
- Wait for the agent to complete and capture the temperature value from its report
|
|
```
|
|
|
|
**Why it failed:** Too vague. The agent interpreted "launch" as running a bash command instead of using the Task tool properly.
|
|
|
|
#### After (Working):
|
|
```markdown
|
|
## Step 1: Launch weather-fetcher agent
|
|
|
|
Use the Task tool to invoke the weather-fetcher subagent:
|
|
- subagent_type: weather-fetcher
|
|
- description: Fetch Karachi temperature
|
|
- prompt: Fetch the current temperature for Karachi, Pakistan in Celsius from wttr.in API. Return the numeric temperature value in your final report.
|
|
- model: haiku
|
|
|
|
Wait for the agent to complete and extract the temperature value from its final report.
|
|
|
|
## Critical Requirements
|
|
|
|
1. **Use Task Tool Only**: DO NOT use bash commands or any other tools. You must use the Task tool to invoke sub-agents.
|
|
```
|
|
|
|
**Why it works:**
|
|
- Explicitly lists all Task tool parameters
|
|
- Clearly states NOT to use bash commands
|
|
- Provides specific parameter values
|
|
|
|
### Testing the Fix
|
|
|
|
After updating the orchestrator agent definition, test it by invoking the orchestrator:
|
|
|
|
```bash
|
|
# Via slash command
|
|
/weather-karachi
|
|
|
|
# Or directly via Task tool
|
|
Task(subagent_type="weather-orchestrator", description="Run weather workflow", prompt="Orchestrate the complete weather workflow", model="haiku")
|
|
```
|
|
|
|
The orchestrator should now:
|
|
1. Successfully invoke weather-fetcher using the Task tool
|
|
2. Extract the temperature from the fetcher's report
|
|
3. Invoke weather-transformer with the temperature value
|
|
4. Report the complete workflow results
|
|
|
|
### Key Takeaways
|
|
|
|
1. **Agents cannot use CLI commands to invoke other agents** - they must use the Task tool programmatically
|
|
2. **Be explicit with tool usage** - clearly state which tool to use and which tools NOT to use
|
|
3. **Provide complete parameter specifications** - list all required parameters with example values
|
|
4. **Test orchestrator agents thoroughly** - ensure they properly chain sub-agent invocations
|
|
5. **Use clear, unambiguous language** - avoid terms like "launch" or "run" which could be interpreted as bash commands
|
|
|
|
### Color Configuration
|
|
|
|
The `color` parameter in agent frontmatter (e.g., `color: green`) controls the color of the agent's output in the CLI, making it easier to visually distinguish between different agents' outputs. This is purely a display feature and does not affect the agent's functionality or the content it produces.
|