187 lines
4.8 KiB
Markdown
187 lines
4.8 KiB
Markdown
---
|
|
name: team-consult
|
|
description: Consult with the full team on any topic - automatically routes to relevant agents
|
|
tools: Read, Write, Glob, Grep, Bash, TodoWrite, AskUserQuestion, Task
|
|
model: inherit
|
|
arguments:
|
|
- name: question
|
|
description: The question or task to consult the team about
|
|
required: true
|
|
---
|
|
|
|
# Team Consult
|
|
|
|
Consult with your virtual webapp team on any topic. Questions are automatically routed to the most relevant specialists who provide their perspectives.
|
|
|
|
## Instructions
|
|
|
|
### Step 1: Analyze Input
|
|
|
|
Parse `$ARGUMENTS.question` to determine relevant agents. Look for:
|
|
|
|
**Technical Implementation Keywords**:
|
|
- code, implement, build, develop, architecture, refactor → `full-stack-engineer`
|
|
- frontend, react, component, CSS, UI → `frontend-engineer`
|
|
- backend, API, database, server → `backend-engineer`
|
|
- deploy, CI/CD, infrastructure → `devops-engineer`
|
|
|
|
**User Experience Keywords**:
|
|
- design, UX, flow, accessibility, user → `ui-ux-designer`
|
|
- requirements, story, prioritize, roadmap → `product-manager`
|
|
|
|
**Security/Quality Keywords**:
|
|
- security, auth, vulnerability, OWASP → `security-engineer`
|
|
- test, QA, bug, quality → `qa-engineer`
|
|
|
|
**Growth/Data Keywords**:
|
|
- growth, marketing, SEO, conversion → `growth-marketer`
|
|
- analytics, metrics, data, tracking → `data-analyst`
|
|
|
|
**Content/Support Keywords**:
|
|
- content, copy, writing → `content-creator`
|
|
- user feedback, support, documentation → `customer-support`
|
|
|
|
### Step 2: Check for @Mentions
|
|
|
|
If the user includes @mentions, prioritize those agents:
|
|
- `@full-stack` → `full-stack-engineer`
|
|
- `@frontend` → `frontend-engineer`
|
|
- `@backend` → `backend-engineer`
|
|
- `@devops` → `devops-engineer`
|
|
- `@pm` or `@product` → `product-manager`
|
|
- `@design` or `@ux` → `ui-ux-designer`
|
|
- `@security` → `security-engineer`
|
|
- `@qa` → `qa-engineer`
|
|
- `@growth` → `growth-marketer`
|
|
- `@data` → `data-analyst`
|
|
- `@content` → `content-creator`
|
|
- `@support` → `customer-support`
|
|
|
|
### Step 3: Select Agents (Max 4)
|
|
|
|
Choose 3-4 most relevant agents. Order by:
|
|
1. Strategy/requirements first (PM, UX)
|
|
2. Implementation second (Engineers)
|
|
3. Support functions third (QA, Security, Growth)
|
|
|
|
Avoid including too many agents - that creates noise.
|
|
|
|
### Step 4: Invoke Agents
|
|
|
|
For each selected agent, use the Task tool to invoke them with:
|
|
|
|
```
|
|
Context: [Original user question]
|
|
|
|
Provide your perspective as the [role] on this question. Focus on:
|
|
- Key considerations from your domain
|
|
- Recommendations
|
|
- Potential concerns or risks
|
|
- Questions that need answering
|
|
|
|
Keep your response concise (3-5 key points).
|
|
```
|
|
|
|
### Step 5: Synthesize Responses
|
|
|
|
Compile agent responses into a structured summary:
|
|
|
|
```markdown
|
|
## Team Consultation: [Topic]
|
|
|
|
### Consulted Specialists
|
|
- [Agent 1] - [Role]
|
|
- [Agent 2] - [Role]
|
|
- [Agent 3] - [Role]
|
|
|
|
---
|
|
|
|
### Consensus Points
|
|
Areas where the team agrees:
|
|
- [Point 1]
|
|
- [Point 2]
|
|
- [Point 3]
|
|
|
|
### Key Perspectives
|
|
|
|
#### [Agent 1 Role]
|
|
[Summary of their input]
|
|
|
|
**Recommendations**:
|
|
- [Recommendation 1]
|
|
- [Recommendation 2]
|
|
|
|
#### [Agent 2 Role]
|
|
[Summary of their input]
|
|
|
|
**Recommendations**:
|
|
- [Recommendation 1]
|
|
- [Recommendation 2]
|
|
|
|
[Continue for each agent...]
|
|
|
|
---
|
|
|
|
### Trade-offs Identified
|
|
| Option | Pros | Cons | Recommended By |
|
|
|--------|------|------|----------------|
|
|
| [Option A] | [Pros] | [Cons] | [Agent] |
|
|
| [Option B] | [Pros] | [Cons] | [Agent] |
|
|
|
|
### Open Questions
|
|
Questions that need resolution:
|
|
- [ ] [Question 1] - Ask: [Who to ask]
|
|
- [ ] [Question 2] - Ask: [Who to ask]
|
|
|
|
### Dissenting Opinions
|
|
If agents disagree:
|
|
- **[Agent 1]** thinks [X] because [reason]
|
|
- **[Agent 2]** thinks [Y] because [reason]
|
|
|
|
---
|
|
|
|
### Recommended Next Steps
|
|
Based on team input:
|
|
1. [Action 1]
|
|
2. [Action 2]
|
|
3. [Action 3]
|
|
|
|
### Additional Consultation Needed?
|
|
[Suggest which other agents might have valuable input if any]
|
|
```
|
|
|
|
## Examples
|
|
|
|
### Example 1: Technical Question
|
|
**Input**: "How should we implement user authentication?"
|
|
|
|
**Agents Selected**:
|
|
1. Security Engineer (security considerations)
|
|
2. Full-Stack Engineer (implementation approach)
|
|
3. Backend Engineer (API/session design)
|
|
4. Product Manager (requirements/user impact)
|
|
|
|
### Example 2: Feature Question
|
|
**Input**: "Should we add dark mode to the app?"
|
|
|
|
**Agents Selected**:
|
|
1. Product Manager (prioritization, user value)
|
|
2. UI/UX Designer (design system impact)
|
|
3. Frontend Engineer (implementation complexity)
|
|
4. Growth Marketer (adoption metrics)
|
|
|
|
### Example 3: With @Mentions
|
|
**Input**: "@security @devops How should we handle secrets in CI/CD?"
|
|
|
|
**Agents Selected**:
|
|
1. Security Engineer (mentioned)
|
|
2. DevOps Engineer (mentioned)
|
|
3. Full-Stack Engineer (general implementation context)
|
|
|
|
## Notes
|
|
|
|
- Keep agent responses focused (3-5 key points each)
|
|
- Highlight disagreements - they're valuable
|
|
- Suggest follow-up with specific agents if needed
|
|
- Don't overload with too many perspectives
|