handoff

handoff is a command for Claude Code from khanh-vu/claude-force. It costs 6 tokens per session (3,367 once invoked), scanned A, original, MIT.

A command that creates a written summary of the current coding session for someone who will continue the work later. It records progress, reasons for decisions, remaining priorities, and the next steps in a saved file.

In plain words
What is it for?
Use it at breaks, phase completions, task switches, or team handoffs to document what happened and what remains.
Why use it?
Important context can be lost when a long session ends, the context limit is reached, or work moves to another person. The handoff keeps the unfinished work understandable.

Command for Claude Code

Written for Claude Code: a Claude Code command (commands/*.md). Also seen: positional $N argument.

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/khanh-vu/claude-force/handoff
Clone the repo
git clone --depth 1 https://github.com/khanh-vu/claude-force

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for handoff

README.md
[![agentmods](https://agentmods.dev/badge/commands/khanh-vu/claude-force/handoff.svg)](https://agentmods.dev/commands/khanh-vu/claude-force/handoff)
Your own site
<a href="https://agentmods.dev/commands/khanh-vu/claude-force/handoff"><img src="https://agentmods.dev/badge/commands/khanh-vu/claude-force/handoff.svg" alt="Measured on agentmods" height="20"></a>
Per session 6 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 3,367 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00006 $0.03367
Opus 5 $0.00003 $0.01684
Sonnet 5 $0.00001 $0.00673
Haiku 4.5 $0.00001 $0.00337

Measured 5d ago against content hash d1d94052fd76, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade A, and why

handoff scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

claude_force/templates/commands/handoff.md · 588 lines

How it starts

The opening of the file, as written. The whole thing — 588 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Handoff Command

Create structured handoff documentation for seamless session continuation.

When to Use

Perfect for:

  • End of work session (before closing)
  • Before context window fills up
  • Natural breakpoints (workflow completion)
  • Switching to different task
  • Team handoffs (sharing context)

Timing:

  • Sessions > 2 hours: Recommended
  • Before major breaks: Essential
  • Complex workflows: Every phase completion

Basic Usage

Generate handoff for current session:

/handoff

What happens:

  1. Analyzes current session state
  2. Extracts workflow progress
  3. Captures decision context (WHY you made choices)
  4. Prioritizes remaining work (P1/P2/P3)
  5. Includes governance and performance metrics
  6. Auto-detects confidence level
  7. Saves to .claude/whats-next.md and archives

Output:

## Session Handoff

**Session**: session-20251116-143022
**Started**: 2025-11-16 14:30
**Duration**: 4h 15m
**Status**: 🟢 High Confidence

---

## Original Task

**Title**: Build Product Catalog UI
**Priority**: 🔴 High
**Workflow**: full-stack-feature

Create responsive product catalog with filters, search, and pagination.
Support 1000+ products with good performance.

---

## Session Summary

**Key Decisions Made:**
- Chose PostgreSQL array types for product tags (better query performance than JSON)
- React Query for API caching with 5min stale time (balance freshness/performance)
- Filter state in URL params for shareable links
- Pagination set to 50 items/page (UX testing showed optimal)

**Critical Insights:**
- Backend API already handles sorting, don't duplicate in frontend
- Product images need lazy loading (performance issue discovered)
- Search debouncing at 300ms feels responsive (tested)

**Conversation Highlights:**
Spent first hour on architecture decisions (API design, state management).
Implemented core components mid-session. Discovered performance issue with
images late in session, added to todos for next session.

---

## Progress Summary

**Overall**: 5 of 8 agents complete (62%)

✅ frontend-architect
✅ database-architect
✅ backend-architect
✅ python-expert
✅ ui-components-expert
🔄 frontend-developer (Next - 60% complete)
⏳ qc-automation-expert
⏳ deployment-integration-expert

---

## Work Completed

**Completed Items:**
- ✅ Designed component architecture
- ✅ Created database schema with indexes
- ✅ Built REST API endpoints (/products, /search, /filter)
- ✅ Implemented ProductCard, FilterBar, SearchBox components

**Files Modified:**
- `src/components/ProductCard.tsx`: Created product card with image, price, tags
- `src/components/FilterBar.tsx`: Filter UI with category, price range, tags
- `src/api/products.py`: REST endpoints with pagination, sorting, filtering
- `database/schema.sql`: Product table with GIN index on tags array

**Agent Outputs:**
- **frontend-architect**: Component tree, state flow diagram, routing structure
- **database-architect**: Schema with indexes, migration scripts
- **backend-architect**: API spec (OpenAPI), caching strategy
- **python-expert**: Optimized queries with SQLAlchemy, bulk loading
- **ui-components-expert**: Responsive components with Tailwind CSS

---

## Next Steps

**PRIORITY 1 (Critical Path):**
- 🔴 Complete ProductList container component (ties everything together)
- 🔴 Implement lazy loading for product images (performance blocker)

**PRIORITY 2 (High Value):**
- 🟡 Add loading states and error boundaries
- 🟡 Implement pagination controls

**PRIORITY 3 (Nice to Have):**
- 🟢 Add sorting options (name, price, date)
- 🟢 Add filter presets (e.g., "On Sale", "New Arrivals")

**Dependencies:**
- QA testing depends on ProductList completion
- Deployment depends on QA passing

---

## Active Context

**Most Relevant Right Now:**
- 💡 ProductList needs virtualization for 1000+ products (react-window library)
- 💡 Images hosted on CDN, use srcset for responsive loading
- 💡 Filter state structure: `{ category: string[], priceRange: [min, max], tags: string[] }`

**Known Blockers:**
- ⚠️ CDN access not yet configured for images
  - Mitigation: Use placeholder images for development, add todo for ops

**Open Questions:**
- ❓ Should we pre-load next page of results? (UX vs bandwidth trade-off)
- ❓ Real-time stock updates or polling? (need product owner input)

---

## Quality Status

**Last Validation**: ✅ All Checks Pass

- Scorecard: 12/12 ✅
- Write Zone: Updated ✅
- Secrets: None detected ✅
- Format: Valid ✅

---

## Cost & Performance

💰 **Total Cost**: $2.45
⏱️ **Execution Time**: 4h 15m
🤖 **Agents Run**: 5 of 8
📊 **Tokens Used**: 45,230 tokens
📈 **Context Window**: 35.2% used

---

## Technical Context

Key decisions and gotchas:
- Using PostgreSQL array types for tags (better than JSONB for this use case)
- React Query configured with `staleTime: 300000` (5 minutes)
- Filter state managed in URL params: `?category=electronics&priceMin=100&priceMax=500`
- API pagination: limit/offset (consider cursor-based if performance issues)
- Image CDN URLs follow pattern: `${CDN_BASE}/products/${productId}/${size}.webp`
- Frontend validation matches backend (shared validation lib recommended)

---

## To Resume

**To Resume This Session:**

1. **Review this handoff** - Read session summary and active context above
2. **Start with**: Complete ProductList container component (PRIORITY 1)
3. **Continue workflow**: `/run-agent frontend-developer` (will resume from 60%)
4. **After frontend**: Run `/run-agent qc-automation-expert` for testing
5. **Validate**: Run `/validate-output` before proceeding to deployment

**Quick Start Command**:

/run-agent frontend-developer


---

**Generated**: 2025-11-16 18:45:22
**Saved to**: `.claude/handoffs/handoff-2025-11-16-184522.md`

Read the full file on GitHub · 588 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 5d ago First seen · 588 lines · 6 tokens per session scan A d1d94052fd76

Subscribe to this mod's changes

handoff is a command published in the GitHub repository khanh-vu/claude-force (5 stars, last pushed 9mo ago), licensed MIT. It adds 6 tokens to every session and 3,367 once invoked, about $0.0000 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.