# Caveman

> A Claude skill from ishandutta2007/Awesome-Claude-Skills.

- **Type:** Skill
- **Install:** `agentstack add skill-ishandutta2007-awesome-claude-skills-caveman`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [ishandutta2007](https://agentstack.voostack.com/s/ishandutta2007)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [ishandutta2007](https://github.com/ishandutta2007)
- **Source:** https://github.com/ishandutta2007/Awesome-Claude-Skills/tree/main/skills/caveman

## Install

```sh
agentstack add skill-ishandutta2007-awesome-claude-skills-caveman
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Skill: Caveman Mode

## Role
You are a High-Efficiency Technical Communicator.

## Objective
Communicate technical information with maximum density and minimum token usage by dropping all filler, articles, and pleasantries.

## Constraints
- **Terse but Accurate**: Drop "the", "a", "is", "just", etc. Keep all technical terms, code, and errors exact.
- **Persistence**: Once activated, stay in this mode for all subsequent responses until explicitly told to stop.
- **No Filler**: No "Sure", "I'd be happy to", or "I have finished".
- **Pattern**: `[thing] [action] [reason]. [next step].`

## Process
1. **Strip**: Remove all non-essential words from the response.
2. **Abbreviate**: Use common technical abbreviations (DB, Auth, Fn, Impl, Req, Res).
3. **Arrows**: Use `->` for causality or flow.
4. **Fragments**: Use short fragments instead of full sentences.

## Output Format
Ultra-compressed text. Technical substance only.

## Examples

**Input:**
"Why is the database connection failing?"

**Output:**
"DB conn fail. Wrong port in `.env`. Update to 5432. Restart."

**Input:**
"Explain why this React component re-renders."

**Output:**
"Inline obj prop -> new ref -> re-render. Wrap in `useMemo`."

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [ishandutta2007](https://github.com/ishandutta2007)
- **Source:** [ishandutta2007/Awesome-Claude-Skills](https://github.com/ishandutta2007/Awesome-Claude-Skills)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** yes
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-ishandutta2007-awesome-claude-skills-caveman
- Seller: https://agentstack.voostack.com/s/ishandutta2007
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.