agent-handoff-protocols
Use this skill when designing task handoffs between agents. Activate when the user needs to pass work between agents, transfer context between agents, implement agent-to-agent communication, or design protocols for agents to collaborate on sequential tasks.
What this skill does
# Agent Handoff Protocols
Design clean, reliable protocols for passing tasks and context between AI agents.
## When to Use
- Agents need to pass work to each other
- Context must be preserved across agent transitions
- Building agent pipelines or workflows
- Implementing specialized agent chains
## Handoff Fundamentals
### What Gets Transferred
```typescript
interface HandoffPackage {
// The work item
task: {
id: string;
type: string;
description: string;
requirements: string[];
};
// Context from previous work
context: {
originalRequest: string;
completedSteps: Step[];
intermediateResults: Record<string, unknown>;
relevantFiles: string[];
decisions: Decision[];
};
// Metadata
metadata: {
sourceAgent: string;
targetAgent: string;
timestamp: Date;
priority: number;
deadline?: Date;
attempt: number;
};
// Expectations
expectations: {
outputFormat: JSONSchema;
successCriteria: string[];
timeoutMs: number;
};
}
```
### Handoff States
```
┌─────────┐ ┌──────────┐ ┌──────────┐ ┌──────────┐
│ PENDING │───►│ ACCEPTED │───►│ WORKING │───►│ COMPLETE │
└─────────┘ └──────────┘ └──────────┘ └──────────┘
│ │ │ │
▼ ▼ ▼ ▼
┌─────────┐ ┌──────────┐ ┌──────────┐ ┌──────────┐
│ TIMEOUT │ │ REJECTED │ │ FAILED │ │ RETURNED │
└─────────┘ └──────────┘ └──────────┘ └──────────┘
```
## Protocol Patterns
### Pattern 1: Direct Handoff
Simple point-to-point transfer.
```typescript
class DirectHandoff {
async transfer(
from: Agent,
to: Agent,
package: HandoffPackage
): Promise<HandoffResult> {
// Validate target can accept
const canAccept = await to.canAccept(package);
if (!canAccept.ok) {
return { status: 'rejected', reason: canAccept.reason };
}
// Prepare context summary
const contextSummary = await from.summarizeContext(package.context);
// Execute handoff
const ack = await to.receive({
...package,
context: {
...package.context,
summary: contextSummary
}
});
if (ack.accepted) {
await from.releaseTask(package.task.id);
return { status: 'accepted', receivedAt: ack.timestamp };
}
return { status: 'rejected', reason: ack.reason };
}
}
```
### Pattern 2: Queued Handoff
For async/decoupled systems.
```typescript
class QueuedHandoff {
constructor(private queue: MessageQueue) {}
async submit(package: HandoffPackage): Promise<string> {
const ticket = generateTicketId();
await this.queue.enqueue({
ticket,
package,
status: 'pending',
submittedAt: new Date()
});
return ticket;
}
async checkStatus(ticket: string): Promise<HandoffStatus> {
return this.queue.getStatus(ticket);
}
// Target agent polls for work
async claim(agentId: string): Promise<HandoffPackage | null> {
const item = await this.queue.dequeue({
filter: { targetAgent: agentId }
});
if (item) {
await this.queue.updateStatus(item.ticket, 'claimed');
}
return item?.package || null;
}
}
```
### Pattern 3: Brokered Handoff
Intermediary manages transfers.
```typescript
class HandoffBroker {
private pending = new Map<string, PendingHandoff>();
async initiateHandoff(
source: Agent,
targetCapability: string,
package: HandoffPackage
): Promise<HandoffResult> {
// Find suitable targets
const candidates = await this.registry.findByCapability(targetCapability);
// Negotiate with candidates
for (const candidate of candidates) {
const offer = await candidate.negotiate({
taskType: package.task.type,
estimatedTokens: this.estimateTokens(package),
deadline: package.metadata.deadline
});
if (offer.accepted) {
// Execute transfer through broker
return this.executeTransfer(source, candidate, package, offer);
}
}
// No takers - queue or return to source
return this.handleNoTakers(source, package);
}
}
```
## Context Transfer Strategies
### Strategy 1: Full Context
Transfer everything - simple but expensive.
```typescript
function fullContextTransfer(context: Context): TransferredContext {
return {
...context,
_transferType: 'full',
_tokenEstimate: estimateTokens(context)
};
}
```
### Strategy 2: Summarized Context
AI summarizes before transfer.
```typescript
async function summarizedTransfer(
context: Context,
targetNeeds: string[]
): Promise<TransferredContext> {
const summary = await summarizeAgent.run(`
Summarize this context for an agent that needs to:
${targetNeeds.join('\n')}
Context:
${JSON.stringify(context)}
Keep only information relevant to those needs.
Be concise but complete.
`);
return {
summary,
_transferType: 'summarized',
_originalTokens: estimateTokens(context),
_summaryTokens: estimateTokens(summary)
};
}
```
### Strategy 3: Reference-Based
Transfer references, target fetches as needed.
```typescript
interface ReferenceContext {
taskId: string;
contextUrl: string; // URL to fetch full context
keyPoints: string[]; // Essential facts
fileReferences: {
path: string;
relevantSections: string[];
}[];
}
async function referenceTransfer(context: Context): Promise<ReferenceContext> {
// Store full context
const contextId = await contextStore.save(context);
// Extract key points
const keyPoints = await extractKeyPoints(context);
return {
taskId: context.taskId,
contextUrl: `context://${contextId}`,
keyPoints,
fileReferences: context.relevantFiles.map(f => ({
path: f.path,
relevantSections: f.sections
}))
};
}
```
## Handoff Instructions Template
Include clear instructions for receiving agent.
```markdown
## Handoff Package
### Task
[Clear description of what needs to be done]
### Context Summary
- Original request: [What the user asked for]
- What's been done: [Completed steps]
- Current state: [Where we are now]
### Your Specific Task
[Exact instructions for this agent]
### Key Information
- [Critical fact 1]
- [Critical fact 2]
- [Decision that was made and why]
### Constraints
- [Time/token budget]
- [Must use existing patterns in X file]
- [Cannot modify Y]
### Expected Output
[Format and content expectations]
### Success Criteria
- [Criterion 1]
- [Criterion 2]
### If You Get Stuck
- [Fallback option 1]
- [Escalation path]
```
## Error Handling
### Rejection Handling
```typescript
async function handleRejection(
package: HandoffPackage,
reason: string
): Promise<HandoffResult> {
// Log rejection
await log.handoffRejected(package.task.id, reason);
if (package.metadata.attempt < MAX_ATTEMPTS) {
// Retry with different target
return retryHandoff({
...package,
metadata: {
...package.metadata,
attempt: package.metadata.attempt + 1
}
}, { excludeAgent: package.metadata.targetAgent });
}
// Max retries - escalate
return escalateToHuman(package, `Handoff failed: ${reason}`);
}
```
### Timeout Handling
```typescript
async function monitorHandoff(
ticket: string,
timeoutMs: number
): Promise<void> {
const deadline = Date.now() + timeoutMs;
while (Date.now() < deadline) {
const status = await handoffQueue.getStatus(ticket);
if (status === 'complete') return;
if (status === 'failed') throw new HandoffError('Task failed');
await sleep(POLL_INTERVAL);
}
// Timeout
await handoffQueue.cancel(ticket);
throw new HandoffTimeoutError(ticket);
}
```
### Rollback Handling
```typescript
async function rollbackHandoff(
package: HandoffPackage,
reason: string
): Promise<void> {
// Return task to source
await package.metadata.sourceAgent.receive({
...package,
context: {
...package.context,
Related in Design
contribute
IncludedLocal-only OSS contribution command center. Auto-refreshes the user's in-flight PR and issue state on invoke so conversations start with full context — no need to brief Claude on what's in flight. Helps the user find issues to contribute to on GitHub, builds per-repo dossiers of what each upstream expects (CLA, DCO, branch convention, AI policy, draft-first, review bots, issue templates), runs deterministic gates before any external action so AI-assisted contributions don't reach maintainers as slop. State is markdown-only: candidate files at ~/.contribute-system/candidates/, repo dossiers at ~/.contribute-system/research/, append-only event log at ~/.contribute-system/log.jsonl. No database, no cloud calls. Use when the user asks about their PRs / issues / contributions, wants to find new work to take on, claim an issue, build/refresh a repo's dossier, or draft a Design Issue or PR. Trigger with "/contribute", "what's my PR status", "find a contribution", "claim issue X", "draft a Design Issue for Y", "refresh dossier for Z".
architectural-analysis
IncludedUser-triggered deep architectural analysis of a codebase or scoped subtree across eight modes — information architecture, data flow, integration points, UI surfaces, interaction patterns, data model, control flow, and failure modes. This skill should be used when the user asks to "diagram this codebase," "map the architecture," "show the data flow," "give me an ERD," "trace control flow," "find the integration points," "verify the layout pattern," "audit the UX architecture," or any similar request whose primary deliverable is mermaid diagrams plus cited reports under docs/architecture/. Dispatches haiku/sonnet sub-agents in parallel for per-mode exploration, then verifies every citation mechanically before any node lands in a diagram. Not for one-off prose explanations of code (use code-explanation) or for high-level system design from scratch (use system-design).
mcp
IncludedModel Context Protocol (MCP) server development and tool management. Languages: Python, TypeScript. Capabilities: build MCP servers, integrate external APIs, discover/execute MCP tools, manage multi-server configs, design agent-centric tools. Actions: create, build, integrate, discover, execute, configure MCP servers/tools. Keywords: MCP, Model Context Protocol, MCP server, MCP tool, stdio transport, SSE transport, tool discovery, resource provider, prompt template, external API integration, Gemini CLI MCP, Claude MCP, agent tools, tool execution, server config. Use when: building MCP servers, integrating external APIs as MCP tools, discovering available MCP tools, executing MCP capabilities, configuring multi-server setups, designing tools for AI agents.
react-native-skia
IncludedDesign, build, debug, and optimise high-polish animated graphics in React Native or Expo using @shopify/react-native-skia, Reanimated, and Gesture Handler. Use when the user wants canvas-driven UI, shaders, paths, rich text, image filters, sprite fields, Skottie, video frames, snapshots, web CanvasKit setup, or performance tuning for custom motion-heavy elements such as loaders, hero art, cards, charts, progress indicators, particle systems, or gesture-driven surfaces. Also use when the user asks for fluid, glow, glass, blob, parallax, 60fps/120fps, or GPU-friendly animated effects in React Native, even if they do not explicitly say "Skia". Do not use for ordinary form/layout work with standard views.
plaid
IncludedProduct Led AI Development — guides founders from idea to launched product. Six capabilities: Idea (discover a product idea), Validate (pressure-test the idea against fatal flaws, problem reality, competition, and 2-week MVP feasibility), Plan (vision intake + document generation), Design (translate image references into a design.md spec), Launch (go-to-market strategy), and Build (roadmap execution). Use when someone says "PLAID", "plaid idea", "help me find an idea", "product idea", "idea from my business", "idea from my expertise", "plaid validate", "validate my idea", "pressure-test", "is this idea good", "find fatal flaws", "validate the problem", "plan a product", "define my vision", "generate a PRD", "product strategy", "plaid design", "design from image", "translate image to design", "create design.md", "extract design tokens", "plaid launch", "go-to-market", "launch plan", "GTM strategy", "launch playbook", "plaid build", "build the app", "start building", or "execute the roadmap".
nextjs-framer-motion-animations
IncludedAdds production-safe Motion for React or Framer Motion animations to Next.js apps, including reveal, hover and tap micro-interactions, whileInView, stagger, AnimatePresence, layout and layoutId transitions, reorder, scroll-linked UI, and lightweight route-content transitions. Use when the user asks to add, refactor, or debug Motion or Framer Motion in App Router or Pages Router codebases, especially around server/client boundaries, reduced motion, LazyMotion, bundle size, hydration, or route transitions. Avoid for GSAP-style timelines, WebGL or 3D scenes, heavy scroll storytelling, or CSS-only effects unless Motion is explicitly requested.