REMvisual/claude-handoff / skills/handoff
handoff
VerifiedCreate a structured session handoff when context is running low or work is pausing. Deep context mining, self-validation, multi-file splitting. Captures everything the next session needs.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 12.8 KB |
| references/close-session.md | markdown | 3.2 KB |
| references/mining-deep-chunked.md | markdown | 2.9 KB |
| references/output-template.md | markdown | 4.7 KB |
| references/validation.md | markdown | 2.1 KB |
references/validation.md
View on GitHub# Handoff Self-Validation Checklist
Run after Step 4 Write. Do not skip.
## 1. Line Count Check
Check your system prompt for context window size. Target the CEILING, not the floor.
| Pass | Minimum | Target ceiling | Must expand if under |
|---|---:|---:|---:|
| Quick (Standard/200K) | 150 | 400 | 150 |
| Quick (Extended/1M) | 250 | 800 | 250 |
| Deep | 300 | 600 | 300 |
| Chunked | 500 | 800 | 500 |
**Under MUST-expand threshold:** Run Phase 2 (gap research). Read your file back. Scan conversation for uncaptured data. Use Edit to append. Do NOT proceed until above threshold.
**Between threshold and ceiling:** Phase 2 still recommended. There's almost certainly data you missed.
Common thin-section culprits:
- "Where We Are" has <10 bullets
- "What We Tried" missing or has only 1-2 entries
- "Evidence & Data" summarizes instead of giving numbers
- "Key Decisions" has only 1 entry
- "Code Analysis" is missing when source was read during the session
## 2. Data Completeness Check
- [ ] "Where We Are" includes specific file AND function names
- [ ] "What We Tried" has one entry per distinct approach discussed
- [ ] "Evidence & Data" has actual numbers, not summaries ("error rate: 28.6" not "high error")
- [ ] "Key Decisions" includes at least one rejected alternative
- [ ] If prior handoffs exist on this topic: clear "what changed since last time"
- [ ] "Quick Start" has a concrete first action, not "continue working"
- [ ] Data file paths included so next session can reference raw results
## 3. Chain Check
- [ ] **Chain** line has a valid tag (epic, bead ID(s), or standalone hex)
- [ ] If continuation: **Parent** file actually exists (ls to verify)
- [ ] **Prior chain** breadcrumb lists all ancestors in order
- [ ] If seq 1: Parent = `none — first in chain`
- [ ] Parent is NOT an auto-handoff (file header doesn't contain `**Auto:** true`)
## 4. Split Check
Over the split threshold (400 standard / 800 extended)? **SPLIT** into part1 + part2 with cross-references.
## 5. If any check fails
Fix before proceeding to Steps 5+. Rewrite thin sections. You have ~25% context remaining — use it.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack." | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |