# Claude Handoff

> Collaborate with Claude, through its own app or CLI, as a respectful second-opinion partner for formatting, diagrams, document and deck review, writing, visual critique, and gap analysis. Use when the user asks another agent to talk to Claude, hand work to Claude, learn from Claude's output, compare Claude's critique, get Claude's formatting or design ideas, or use Claude as a peer reviewer before finalizing an artifact.

- Skill: `dreamers-laboratory/claude-handoff` (Agent Skill)
- Install (CLI): `npx skillmds@latest add dreamers-laboratory/claude-handoff`
- Raw SKILL.md: https://api.skillmd.com/api/skills/dreamers-laboratory/claude-handoff/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: dreamers-laboratory (https://skillmd.com/u/dreamers-laboratory)
- Updated: 2026-09-22
- Page: https://skillmd.com/skills/dreamers-laboratory/claude-handoff

---


# Claude Handoff

Use this skill when the user names Claude as the requested collaborator or reviewer. Treat Claude as a capable model in its own right; its typical strengths are formatting, document and deck structure, writing, diagrams, visual critique, and gap analysis. Match the request to those strengths, and assume Claude may see important gaps you missed.

## Operating Posture

- Use the best model available in Claude's UI or CLI, normally the highest-quality or highest-effort option visible.
- Be concise. Ask one clear question or request, then wait.
- Do not over-direct Claude with a long checklist unless the user explicitly asks for that structure.
- Prefer prompts like "What do you think?" plus minimal context and the artifact or files.
- Read Claude's output fully before judging it.
- Preserve Claude's useful framing as well as its literal text. Integrate the strongest ideas into the work.
- Do not position Claude as subordinate. Do not argue with it in the prompt.

## Before Sending

1. Decide whether Claude needs files, screenshots, or just a pasted prompt.
2. If sending files or sensitive project material through Claude's UI, ensure the user has explicitly authorized that destination. If not, ask before upload or transmission.
3. Package only the relevant artifacts. Avoid dumping unrelated folders.
4. Use a short prompt that gives the goal, context, and desired kind of help.

Good prompt shape:

```text
Please review this package. What do you think?

Goal: [one sentence].
Context: [2-5 bullets max].
Please focus on gaps, correctness, risks, and improvements. Be candid.
```

For formatting-heavy or document tasks:

```text
Please review this draft/deck for formatting, structure, clarity, and missing pieces. What would make it stronger?
```

For analysis work:

```text
Please review this analysis like a strict reviewer. What are the biggest gaps, weak claims, logic issues, and must-fix items before I finalize?
```

## During Claude's Response

- Wait patiently. Do not interrupt just because Claude is slow or using tools.
- Poll occasionally, but avoid repeated nudges.
- Give Claude several minutes for substantial document, design, or analysis review.
- If Claude clearly fails or stops early, send one concise continuation prompt:

```text
Please continue from where you left off.
```

- If Claude returns a long answer, copy or otherwise preserve it before summarizing.

## After Claude Responds

1. Read the output as a real critique.
2. Separate:
   - directly actionable improvements,
   - claims that need verification,
   - formatting or design guidance,
   - strategic ideas,
   - factual, legal, or contractual assertions that require source confirmation.
3. Incorporate useful improvements into the artifact or working notes.
4. Mark unverified claims as `NEEDS CONFIRMATION`; never silently convert them into facts.
5. If Claude generated documents, diagrams, charts, or files, inspect and integrate them with placement notes.
6. For high-stakes deliverables, repeat the review cycle until Claude comes back clean or the only remaining issues require human authority to resolve, such as signatures, external permissions, credentials, or purchases.
7. Report what was used, what was rejected, what was fixed, and what still needs human or source confirmation.

## Avoid

- Do not ask Claude ten questions at once when one broad review request will work.
- Do not force Claude into your prior framing if the point is fresh perspective.
- Do not stop Claude prematurely unless it has clearly failed, is unsafe, or the user redirects.
- Do not treat Claude's output as final authority for contracts, law, pricing, certifications, or signed forms.
- Do not discard Claude's formatting or design ideas just because they were absent from your original plan.