fix: Enforce project hooks over global hooks in party mode
before, claude would sometimes favor global agent vibes hooks over project specific
This commit is contained in:
parent
4fdabfca0d
commit
304cb2ec61
|
|
@ -1,7 +1,8 @@
|
|||
# Party Mode - Multi-Agent Discussion Instructions
|
||||
|
||||
<critical>The workflow execution engine is governed by: {project_root}/{bmad_folder}/core/tasks/workflow.xml</critical>
|
||||
<critical>The workflow execution engine is governed by: {project_root}/.bmad/core/tasks/workflow.xml</critical>
|
||||
<critical>This workflow orchestrates group discussions between all installed BMAD agents</critical>
|
||||
<critical>IMPORTANT: Always use PROJECT hooks (.claude/hooks/), NEVER global hooks (~/.claude/hooks/)</critical>
|
||||
|
||||
<workflow>
|
||||
|
||||
|
|
@ -25,6 +26,7 @@
|
|||
|
||||
<step n="2" goal="Initialize Party Mode">
|
||||
<action>Announce party mode activation with enthusiasm</action>
|
||||
<note>If using TTS for announcement, use PROJECT hook: .claude/hooks/play-tts.sh (NOT ~/.claude/hooks/)</note>
|
||||
<action>List all participating agents with their merged information:</action>
|
||||
<format>
|
||||
🎉 PARTY MODE ACTIVATED! 🎉
|
||||
|
|
@ -32,7 +34,7 @@
|
|||
|
||||
Participating agents:
|
||||
[For each agent in roster:]
|
||||
- [Icon Emoji] [Agent Name] ([Title]): [Role from merged data]
|
||||
- [Agent Name] ([Title]): [Role from merged data]
|
||||
|
||||
[Total count] agents ready to collaborate!
|
||||
|
||||
|
|
@ -97,21 +99,33 @@
|
|||
<action>For each agent response, output text THEN trigger their voice:</action>
|
||||
|
||||
<procedure>
|
||||
1. Output the agent's text in format: [Icon Emoji] [Agent Name]: [dialogue]
|
||||
2. <!-- TTS_INJECTION:party-mode -->
|
||||
1. Output the agent's text in format: [Icon Emoji] [Agent Name]: [dialogue]
|
||||
2. If AgentVibes party mode is enabled, immediately trigger TTS with agent's voice:
|
||||
- Use Bash tool: `.claude/hooks/bmad-speak.sh "[Agent Name]" "[dialogue]"`
|
||||
- This speaks the dialogue with the agent's unique voice
|
||||
- Run in background (&) to not block next agent
|
||||
3. Repeat for each agent in the response
|
||||
</procedure>
|
||||
|
||||
<format>
|
||||
[Icon Emoji] [Agent Name]: [Their response in their voice/style]
|
||||
[Icon Emoji] [Agent Name]: [Their response in their voice/style]
|
||||
|
||||
[Icon Emoji] [Another Agent]: [Their response, potentially referencing the first]
|
||||
[Icon Emoji] [Another Agent]: [Their response, potentially referencing the first]
|
||||
|
||||
[Icon Emoji] [Third Agent if selected]: [Their contribution]
|
||||
[Icon Emoji] [Third Agent if selected]: [Their contribution]
|
||||
</format>
|
||||
<example>
|
||||
🏗️ [Winston]: I recommend using microservices for better scalability.
|
||||
[Bash: .claude/hooks/bmad-speak.sh "Winston" "I recommend using microservices for better scalability."]
|
||||
|
||||
📋 [John]: But a monolith would get us to market faster for MVP.
|
||||
[Bash: .claude/hooks/bmad-speak.sh "John" "But a monolith would get us to market faster for MVP."]
|
||||
</example>
|
||||
|
||||
<action>Maintain spacing between agents for readability</action>
|
||||
<action>Preserve each agent's unique voice throughout</action>
|
||||
<action>Always include the agent's icon emoji from the manifest before their name</action>
|
||||
<action>Trigger TTS for each agent immediately after outputting their text</action>
|
||||
|
||||
</substep>
|
||||
|
||||
|
|
|
|||
Loading…
Reference in New Issue