# Compress Technical Replies With Caveman Mode

Long assistant replies can waste attention and context budget. This skill makes Claude, Codex, and Claude Code answer tersely while preserving technical meaning.

## Install

```bash
npx skillstore add mattpocock/caveman
```

## Metadata

- Status: approved
- Slug: mattpocock-caveman
- Skillstore revision: r1
- Version status: missing
- Tree hash: 47f33c27a8babad8d5c767c8f88d547c15670bb4b801aa576dba5e503108aa44
- Author: mattpocock
- GitHub username: mattpocock
- License: MIT
- Repository: https://github.com/mattpocock/skills/tree/main/skills/productivity/caveman/
- Ref: b8ca75d2c0a7e7102978993058777d82b8ab2610
- Supported tools: Claude, Codex, Claude Code
- Audit status: complete
- Agent install advisory: allowed
- Manual install advisory: allowed
- Artifact signature: available
- Audit attestation: unavailable
- Human verification: not\_verified
- Risk factors: external\_commands
- Quality score: 77
- Quality tier: bronze
- Public page: https://skillstore.pages.dev/skills/mattpocock-caveman
- Manifest: https://skillstore.pages.dev/api/skills/mattpocock-caveman/manifest

## Capabilities

- Turns assistant responses into terse technical fragments after activation.
- Removes filler words, pleasantries, hedging, and extra transitions.
- Keeps exact technical terms, code blocks, and error messages unchanged.
- Uses arrows and abbreviations for compact cause and effect explanations.
- Temporarily expands wording for security warnings and irreversible actions.

## Use Cases

- Speed Up Debugging Chats: Get compact root-cause notes, fixes, and next steps while working through technical issues.
- Reduce Context Usage: Keep long Claude, Codex, or Claude Code sessions concise when context budget matters.
- Create Brief Technical Summaries: Convert analysis into short, precise updates for pull requests, tickets, or review notes.

## Prompt Templates

### Enable Caveman Mode

```
Use caveman mode for this session. Keep technical accuracy, but remove filler and pleasantries.
```

### Explain Briefly

```
Explain this in caveman mode: [paste concept, error, or design]. Keep key terms exact.
```

### Debug With Minimal Tokens

```
Caveman mode. Diagnose this issue and give likely cause, fix, and next check: [paste details].
```

### Summarize Technical Findings

```
Use caveman mode to summarize these findings for engineers. Include impact, root cause, fix, and risk: [paste notes].
```

## Limitations

- It changes response style only and does not add new tool access.
- It may feel too terse for onboarding, teaching, or stakeholder updates.
- It depends on explicit stop commands to return to normal style.
- It does not shorten code blocks or exact quoted errors.

## Best Practices

- Use it when you need dense technical answers, not broad teaching.
- Ask for normal mode when clarity matters more than brevity.
- Keep safety warnings explicit, even when the rest of the response is terse.

## Anti Patterns

- Do not use it for sensitive stakeholder communication that needs careful tone.
- Do not use it when a beginner needs step-by-step context.
- Do not shorten legal, security, or destructive-action warnings beyond clear meaning.

## Security Audit

- Audited at: 2026-07-06T17:41:38.755\+00:00
- Summary: All five static findings are false positives caused by Markdown inline code and a fenced SQL example in SKILL.md. No prompt injection, exfiltration intent, or executable shell behavior was found in the reviewed file.

## Stats

- Views: 83
- Downloads: 36
- Favorites: 1
- Popularity score: 0