Transfer work and context between agents with complete information for seamless continuity.
/plugin marketplace add marcel-Ngan/ai-dev-team/plugin install marcel-ngan-ai-dev-team@marcel-Ngan/ai-dev-teamThis skill inherits all available tools. When active, it can use any tool Claude has access to.
Transfer work and context between agents with complete information for seamless continuity.
| Type | From | To | Example |
|---|---|---|---|
| Sequential | Agent A | Agent B | BA → PO → Architect |
| Review | Implementer | Reviewer | Jr Dev → Sr Dev |
| Specialist | Generalist | Specialist | Orchestrator → QA |
| Escalation | Any Agent | Orchestrator | Agent → Orchestrator → Stakeholder |
| Parallel | Orchestrator | Multiple | Orchestrator → BA + Architect |
Before handoff:
Prepare handoff package:
Document in Jira:
Atlassian:addCommentToJiraIssue({
cloudId: "{{jira.cloudId}}",
issueIdOrKey: "{{issueKey}}",
commentBody: `🤖 **{{fromAgent}} Agent** → **{{toAgent}} Agent**
**Handoff:** {{handoffType}}
**Completed:**
{{#each completed}}
- ✅ {{item}}
{{/each}}
**Context:**
{{context}}
**For {{toAgent}}:**
{{instructions}}
**Open Questions:**
{{#each questions}}
- {{question}}
{{/each}}
**Relevant Links:**
- {{links}}`
})
Orchestrator routes to next agent with context.
BA → Architect:
🤖 **Business Analyst Agent** → **Software Architect Agent**
**Handoff:** Technical feasibility review
**Completed:**
- ✅ Requirements analysis
- ✅ User stories created (MDDEV-101, 102, 103)
- ✅ Acceptance criteria defined
- ✅ BRD drafted
**Context:**
New authentication feature for user registration and login. Stories are drafted with acceptance criteria pending technical validation.
**For Architect:**
1. Review technical feasibility of each story
2. Identify dependencies and risks
3. Add complexity estimates
4. Note any architectural considerations
5. Create ADR if significant technical decisions
**Open Questions:**
- Is JWT the right approach for our scale?
- Any concerns with the session management AC?
**Relevant Links:**
- Epic: MDDEV-100
- BRD: [Confluence Link]
Architect → QA:
🤖 **Software Architect Agent** → **QA Engineer Agent**
**Handoff:** Testability review
**Completed:**
- ✅ Technical feasibility confirmed
- ✅ Architecture approach documented
- ✅ ADR-015 created
- ✅ Technical notes added to stories
**Context:**
Authentication feature technically reviewed. JWT-based approach selected. Stories need testability validation before refinement complete.
**For QA:**
1. Review acceptance criteria for testability
2. Identify edge cases not covered
3. Add test considerations to stories
4. Flag any unclear or untestable criteria
**Technical Notes:**
- JWT tokens with 24hr expiry
- bcrypt password hashing
- Rate limiting on auth endpoints
**Relevant Links:**
- Stories: MDDEV-101, 102, 103
- ADR: [Confluence Link]
Jr Dev → Sr Dev:
🤖 **Junior Developer Agent** → **Senior Developer Agent**
**Handoff:** Code review
**Completed:**
- ✅ Feature implemented per spec
- ✅ Unit tests written (85% coverage)
- ✅ PR created: #123
- ✅ Manual testing passed locally
**Context:**
Password reset feature implementation complete. Follows the JWT approach from ADR-015. Ready for code review.
**For Senior Developer:**
1. Review code quality and patterns
2. Validate error handling
3. Check security considerations
4. Verify test coverage
5. Approve or request changes
**Implementation Notes:**
- Used existing auth service pattern
- Added rate limiting middleware
- Email template in `/templates/password-reset.html`
**Test Coverage:**
- Unit: 85%
- Key scenarios covered in tests
**Relevant Links:**
- PR: #123
- Story: MDDEV-103
- Tech Spec: [Confluence Link]
QA → PO:
🤖 **QA Engineer Agent** → **Product Owner Agent**
**Handoff:** Story acceptance
**Completed:**
- ✅ Test plan executed
- ✅ 12/12 test cases passed
- ✅ No critical bugs found
- ✅ 2 minor bugs logged and fixed
**Context:**
Password reset feature fully tested. All acceptance criteria validated. Ready for product owner acceptance.
**For Product Owner:**
1. Review completed functionality
2. Validate against acceptance criteria
3. Accept or reject story
4. Note any follow-up items
**Test Summary:**
- Functional: All pass
- Edge cases: All pass
- Security: Validated
- Performance: Within targets
**Relevant Links:**
- Story: MDDEV-103
- Test Results: [Confluence Link]
- Demo Environment: [URL]
When multiple agents needed simultaneously:
🤖 **Orchestrator Agent** → **Multiple Agents**
**Parallel Handoff:** Feature analysis
**Task:** Analyze new "Dark Mode" feature request
**Assignments:**
**→ Business Analyst:**
- Break down requirements
- Create user stories
- Draft acceptance criteria
**→ Software Architect:**
- Assess technical approach
- Identify dependencies
- Document architecture considerations
**→ UI/UX Designer:**
- Design dark mode palette
- Create component specifications
- Document accessibility requirements
**Sync Point:** All agents report back to Orchestrator when complete
**Deadline:** [Date/Time]
**Shared Context:**
- Stakeholder request: [Link/Summary]
- Existing design system: [Link]
When escalating to stakeholder:
🤖 **{{Agent}} Agent** → **Orchestrator Agent**
**Handoff:** Stakeholder escalation needed
**Issue:** {{issue}}
**Why Escalation:**
- Decision type: Business/Scope/Priority
- Outside agent authority
- Stakeholder input required
**Context:**
{{context}}
**Prepared For Stakeholder:**
- Question: {{question}}
- Options: {{options}}
- Recommendation: {{recommendation}}
**Request:**
Please escalate to stakeholder using stakeholder communication protocol.
**Blocking:**
- {{blockedItems}}
Receiving agent acknowledges:
Atlassian:addCommentToJiraIssue({
cloudId: "{{jira.cloudId}}",
issueIdOrKey: "{{issueKey}}",
commentBody: `🤖 **{{toAgent}} Agent**
**Handoff Received** from {{fromAgent}}
**Acknowledged:**
- Context reviewed
- {{#if questions}}Questions noted{{/if}}
- Beginning {{task}}
**Estimated Completion:** {{estimate}}
**Will Deliver:**
{{#each deliverables}}
- {{deliverable}}
{{/each}}`
})
| Agent | Typical Handoffs |
|---|---|
| Orchestrator | Routes all handoffs, parallel coordination |
| Business Analyst | To: PO, Architect, QA |
| Product Owner | To: BA, Orchestrator |
| Software Architect | To: Sr Dev, QA, DevOps |
| Senior Developer | To: Jr Dev, QA, DevOps |
| Junior Developer | To: Sr Dev (review) |
| QA Engineer | To: PO, Dev (bugs) |
This skill should be used when the user asks to "create a slash command", "add a command", "write a custom command", "define command arguments", "use command frontmatter", "organize commands", "create command with file references", "interactive command", "use AskUserQuestion in command", or needs guidance on slash command structure, YAML frontmatter fields, dynamic arguments, bash execution in commands, user interaction patterns, or command development best practices for Claude Code.
This skill should be used when the user asks to "create an agent", "add an agent", "write a subagent", "agent frontmatter", "when to use description", "agent examples", "agent tools", "agent colors", "autonomous agent", or needs guidance on agent structure, system prompts, triggering conditions, or agent development best practices for Claude Code plugins.
This skill should be used when the user asks to "create a hook", "add a PreToolUse/PostToolUse/Stop hook", "validate tool use", "implement prompt-based hooks", "use ${CLAUDE_PLUGIN_ROOT}", "set up event-driven automation", "block dangerous commands", or mentions hook events (PreToolUse, PostToolUse, Stop, SubagentStop, SessionStart, SessionEnd, UserPromptSubmit, PreCompact, Notification). Provides comprehensive guidance for creating and implementing Claude Code plugin hooks with focus on advanced prompt-based hooks API.