skills.rest
Chris-Maskey
Chris-MaskeyCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

caveman

Compress verbose AI responses into terse technical summaries with configurable intensity modes.

Updated Apr 17, 2026
Search Tags:#conciseness#token-efficiency#developer-communication#caveman-mode#prompt-style#verbosity-control
One-click install
npx skills add https://github.com/Chris-Maskey/opencode-config --skill caveman-chris-maskey
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: caveman
Source: https://github.com/Chris-Maskey/opencode-config/tree/main/skills/caveman
Command: npx skills add https://github.com/Chris-Maskey/opencode-config --skill caveman-chris-maskey

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Many technical AI conversations waste tokens on pleasantries, hedging, and filler which increases cost and reduces signal; caveman compresses language to conserve tokens while preserving technical accuracy and actionability.

Core Features & Use Cases

  • Persistent terse style that reduces token usage by ~75% through removal of filler, articles, and hedging while keeping technical terms intact.
  • Multiple intensity modes (lite, full, ultra and wenyan variants) to control grammatical strictness and compression, auto-trigger on token-efficiency requests, and explicit on/off commands.
  • Use cases: concise bug explanations, compact code review summaries, rapid troubleshooting steps for low-token environments, and terse status reports for logs or changelogs.

Quick Start

Turn on caveman full and summarize the failing auth middleware bug with terse, precise fixes.

Frequently Asked Questions about caveman

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I reduce token usage in AI developer support chats?▼

To reduce token usage in AI developer support chats, use a terse prompt style that strips filler, articles, and hedging while preserving technical accuracy. This compression approach can cut token consumption by roughly 75% without losing code blocks or security warnings.

What's the best way to get concise code review summaries in low-token environments?▼

Concise code review summaries in low-token environments are best generated using a compression mode that removes conversational filler but retains technical terms, irreversible-action confirmations, and code snippets. This ensures compact, actionable feedback for rapid troubleshooting.

Does caveman mode preserve security warnings and irreversible-action confirmations?▼

Yes, caveman mode preserves security warnings and irreversible-action confirmations. While it aggressively removes pleasantries and hedging to conserve tokens, it explicitly safeguards critical code blocks and irreversible-action prompts to maintain technical safety.

Can I control the intensity of verbosity compression for troubleshooting scenarios?▼

You can control verbosity compression intensity using multiple modes like lite, full, ultra, and wenyan variants. These modes adjust grammatical strictness and compression levels, allowing you to tailor terse outputs for specific troubleshooting scenarios or token-efficiency requirements.

Why do my AI technical responses include so much filler and hedging?▼

AI technical responses include filler and hedging due to default verbose settings designed for conversational politeness. Applying a token-efficiency prompt style removes these unnecessary tokens, delivering precise technical content while maintaining actionability for developer tasks.

When should I avoid using terse prompt styles for developer communication?▼

You should avoid using terse prompt styles when context requires detailed explanations, onboarding documentation, or nuanced architectural discussions. While caveman mode excels at bug explanations and status reports, complex technical reasoning may need standard verbosity for clarity.

Related Skills

Explore More Software Engineering Skills→
BrunoAlpezdevBrunoAlpezdev

optimized-session

Apply profile-based lean response constraints to reduce token usage in chat sessions.

Community
Basic
simplefarmer69simplefarmer69
2

devops-bridge

Integrate GitHub, CI/CD, messaging platforms, and issue trackers to automate developer operations.

Community
Advanced
glowingkittyglowingkitty
43

code-walkthrough

Walk through code execution step by step to diagnose logic bugs.

Community
Intermediate
zli5460zli5460
2

mlops/training/trl-fine-tuning

Fine-tune language models on prompt-response datasets with LoRA.

Community
Intermediate