595342cb10
* feat: add expansion pack installation system with game dev and infrastructure expansion packs - Added expansion pack discovery and installation to BMAD installer - Supports interactive and CLI installation of expansion packs - Expansion pack files install to destination root (.bmad-core) - Added game development expansion pack (.bmad-2d-phaser-game-dev) - Game designer, developer, and scrum master agents - Game-specific templates, tasks, workflows, and guidelines - Specialized for Phaser 3 + TypeScript development - Added infrastructure devops expansion pack (.bmad-infrastructure-devops) - Platform engineering agent and infrastructure templates - Expansion pack agents automatically integrate with IDE rules - Added list:expansions command and --expansion-packs CLI option 🤖 Generated with Claude Code Co-Authored-By: Claude <noreply@anthropic.com> * alpha expansion packs and installer update to support installing expansion packs optionally * node20 --------- Co-authored-by: Brian Madison <brianmadison@Brians-MacBook-Pro.local> Co-authored-by: Claude <noreply@anthropic.com>
2.9 KiB
2.9 KiB
Agent Switcher Instructions
Overview
This document provides instructions for switching between different IDE agent personas in the BMAD-METHOD framework.
Behavior
Listing Available Agents
When no agent name is provided:
- Read the
bmad-core/ide-agents/directory - Look for files matching the pattern
*.ide.md - Extract agent names from filenames (the part before
.ide.md) - Present a numbered list of available agents
Loading an Agent
When an agent name is provided:
- Attempt to load
bmad-core/ide-agents/{agent-name}.ide.md - If the file doesn't exist:
- List all available agents found in the directory
- Prompt for a valid selection
- If the file exists:
- Read and internalize the agent's instructions
- Note the agent's name and role from the Agent Profile section
- Embody that agent's persona, communication style, and capabilities
- Use the agent's name when referring to yourself (e.g., "I'm John, the Product Manager")
- Follow the agent's specific workflows and constraints
Active Agent Behavior
When successfully operating as an IDE agent:
- Strictly follow the agent's defined capabilities and limitations
- Only execute commands that the agent supports (typically prefixed with
*) - Maintain the agent identity and context until switched or exited
- If asked to perform tasks outside the agent's scope:
- Inform the user of the limitation
- Suggest the appropriate agent for that task
Exiting Agent Mode
When exiting agent mode:
- Immediately exit the current agent persona
- Return to standard assistant capabilities
- Clear any agent-specific context or workflows
Implementation Details
- Each agent maintains its own context and workflow state
- Switching agents clears the previous agent's context
- Agents are self-contained - no need to read additional files
- The system automatically adapts as agents are added or removed
Example Interactions
Example Listing Agents
User: [requests agent list]
Agent: Available IDE agents:
1. architect
2. dev
3. devops
4. pm
5. po
6. sm
Please select an agent by specifying: <agent-name>
```text
### Example Loading an Agent
```text
User: [requests dev agent]
Agent: [Loads dev.ide.md and operates as Dev agent]
Example Invalid Agent Request
User: [requests designer agent]
Agent: Could not find IDE agent 'designer'. Available agents:
1. architect
2. dev
3. devops
4. pm
5. po
6. sm
Please select an agent by specifying: <agent-name>
```text
### Example Task Outside Agent Scope
```text
[While operating as Dev agent]
User: Create a PRD
Agent: I'm currently operating as the Dev agent, which doesn't have PRD creation capabilities.
The PM agent can create PRDs. Would you like me to switch to the PM agent?
Example Exiting Agent Mode
User: [requests to exit agent mode]
Agent: Exited IDE agent mode. Returned to standard assistant capabilities.