caveman
๐ฏSkillfrom mattpocock/skills
|Ultra-compressed communication mode that reduces token usage by approximately 75% by dropping filler words, articles, and pleasantries while preserving full technical accuracy in responses.
Overview
Part of Matt Pocock's "Skills For Real Engineers" collection, /caveman activates an ultra-compressed communication mode that reduces AI agent token usage by approximately 75%. It achieves this by dropping filler words, articles, pleasantries, and unnecessary transitions while preserving full technical accuracy. The result is terse, information-dense responses that save both time and cost when working with token-metered AI agents.
Key Features
- ~75% token reduction by stripping filler words, articles, and unnecessary language padding from agent responses
- Full technical accuracy preserved so that code suggestions, file paths, error messages, and technical details remain precise and correct
- Instant activation with a single
/cavemancommand that switches the agent's communication style for the rest of the session - Part of a composable collection that works alongside other mattpocock/skills like
/diagnose,/tdd, and/zoom-outwithout interfering with their functionality
Who is this for?
- Developers working with AI coding agents on metered plans who want to minimize token costs without sacrificing technical content quality
- Power users who prefer concise, information-dense responses and find typical AI verbosity wasteful
- Anyone using long-running agent sessions who wants to keep more useful context in the conversation window by reducing output length
Same repository
mattpocock/skills(54 items)
Installation
npx vibeindex add mattpocock/skills --skill cavemannpx skills add mattpocock/skills --skill caveman~/.claude/skills/caveman/SKILL.mdSKILL.md
More from this repository10
Interviews the user relentlessly about every aspect of a plan or design until reaching shared understanding, resolving each branch of the decision tree one by one. Provides recommended answers for each question and explores the codebase when answers can be found there.
A skill that runs a grilling session to challenge your plan against an existing domain model, sharpen terminology by building a shared language in CONTEXT.md, and document architectural decisions in ADRs before coding begins.
An agent skill from Matt Pocock's engineering toolkit that helps improve your codebase architecture. Part of a composable skill set designed to fix common AI coding agent failure modes through structured, experience-based workflows.
Enforces test-driven development with a strict red-green-refactor loop using vertical slices (one test then one implementation at a time), emphasizing behavior verification through public interfaces rather than implementation-coupled testing.
Initial setup skill for Matt Pocock's engineering skills collection that configures your backlog manager, triage labels, and documentation paths for AI coding agents.
A composable skill for triaging issues and tickets using configurable labels, designed to fix common failure modes in AI coding agents by providing structured backlog management workflows.
A collection of small, composable engineering skills designed for real development workflows, featuring grilling sessions for requirement alignment, issue tracking integration, and task triage with customizable labels.
A skill that turns the current conversation context and codebase understanding into a structured PRD (Product Requirements Document) and submits it as a GitHub issue, with modular architecture and user story generation.
Breaks a plan, spec, or PRD into independently-grabbable GitHub issues using tracer-bullet vertical slices. Each issue is a thin end-to-end path through every integration layer that can be demoed or verified on its own.
Composable, model-agnostic agent skills for real engineering workflows, featuring /grill-me for requirement alignment, /prototype for rapid iteration, and task management tools designed to fix common failure modes in AI coding agents.