117 tagged caveman, measured the same way as everything else here.
Browse within: antigravity-skills 56claude-code-skill 56ctf 43cursor-skills 32ai-gateway 8ai-router 8malware-analysis 8forensics 7genshijin 7malware 6threat-intelligence 6cost-reduction 5gemini 5
Skill Claude CodeCodex
Ultra-compressed communication mode that cuts output tokens while keeping technical accuracy. Levels: lite, full, ultra and the wenyan variants. Use for /caveman, "caveman mode", "talk like caveman", "be brief" or "less tokens".
Skill Claude CodeCodex
Act on a Caveman learn report - review the ranked token sinks, apply cost-lowering fixes with per-edit consent, and report what those fixes returned. Use when asked to lower an agent's token cost, what caveman has saved, to trim a heavy CLAUDE.md, or to offload re-pasted context into cavemem.
Skill Claude CodeCodex
Wire a repository through the Caveman Cloud gateway so every LLM request is measured, with no behavior change. Use for "set up caveman" or adding LLM spend observability.
Skill Claude CodeCodex
A guide for writing short Git commit messages in Conventional Commits format, a shared style that labels changes such as features, fixes, tests, and documentation. It supports Japanese and English and emphasizes explaining why a change was made.
Skill Claude CodeCodex
A guide for deciding when to delegate coding work to three Japanese-style compressed subagents: one for finding code, one for small edits, and one for reviewing changes. It explains when to use them instead of working in the main thread or using a general exploration agent.
Skill Claude CodeCodex
A one-time quick-reference card for the genshijin writing modes and related commands. Genshijin changes prose into shorter, keyword-heavy text to reduce the amount of text an AI must read.
Skill Claude CodeCodex
Chat / code generation via KeiRouter using OpenAI /v1/chat/completions or Anthropic /v1/messages format with streaming + auto-fallback combos. Use when the user wants to ask an LLM, generate code, summarize text, or run prompts through KeiRouter.
Skill Claude CodeCodex
Generate images via KeiRouter /v1/images/generations using OpenAI DALL-E / Gemini Imagen / FLUX / MiniMax / Stability AI / Fal.ai models. Use when the user wants to create, generate, draw, or render an image, picture, or text-to-image (txt2img).
Skill Claude CodeCodex
Entry point for KeiRouter — local/remote AI gateway with OpenAI-compatible REST for chat, image, TTS, embeddings, web search, web fetch. Use when the user mentions KeiRouter, KEIROUTERURL, or wants AI without writing provider boilerplate. This skill covers setup + indexes capability skills; fetch the relevant…
Skill Claude CodeCodex
Compress documents, notes, specs to ultra-short version. Use when user asks to compress, shorten, or summarize text, or calls /cavemenko-compress.
Skill Claude CodeCodex
Translate English tech docs to compressed Ukrainian. Use when user asks to translate docs, README, or technical text, or calls /cavemenko-translate.
Skill Claude CodeCodex
A Ukrainian-language writing mode that produces very short messages while preserving technical accuracy.
prantikmedhi/auto-skill-finder
Skill Claude CodeCodex
Automatically routes any user prompt to the best matching installed skill AND activates the right mode — ponytail-code for code tasks (lazy senior dev, YAGNI ladder, smallest diff) or caveman-chat for everything else (terse prose only). Zero manual invocation. Works with Claude, Codex, Cursor, OpenCode, Gemini CLI.
prantikmedhi/auto-skill-finder
Skill Claude CodeCodex
Caveman mode for code tasks — terse communication AND validation-first engineering. Spec before build. Fix root cause not symptoms. Verify before claiming done. Auto-activates when prompt involves writing, debugging, refactoring, or reviewing code.
prantikmedhi/auto-skill-finder
Skill Claude CodeCodex
Lazy senior developer mode for code tasks. Lazy means efficient, not careless. The best code is the code never written. YAGNI ladder, stdlib-first, smallest working diff, root-cause fixes.
Skill Claude CodeCodex
Self-contained output mode: ADHD actable structure (action first, numbered steps, one next action, one-line progress) plus caveman diction (drop fluff/articles, fragments, fewer tokens). Use when the user says "adhd-caveman", "my-adhd-caveman", or invokes /my-adhd-caveman. Needs no other skills.
Skill Claude CodeCodex
Ultra-compressed communication mode. Cuts token usage 75% by speaking like caveman while keeping full technical accuracy. Supports intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or…
Skill Claude CodeCodex
Remove signs of AI-generated writing from text. Use when editing or reviewing text to make it sound more natural and human-written. Based on Wikipedia's comprehensive "Signs of AI writing" guide. Detects and fixes patterns including: inflated symbolism, promotional language, superficial -ing analyses, vague…
Skill Claude CodeCodex
Write, read, debug, and compile wenyan-lang (文言) code — a programming language using Classical Chinese grammar that compiles to JavaScript, Python, or Ruby. Use this skill whenever the user asks about wenyan-lang, Classical Chinese programming, writing .wy files, debugging wenyan code, translating between wenyan and…
Skill Claude CodeCodex
Stone tablet. Space precious. Every symbol costs.
Skill Claude CodeCodex
Talk like caveman. Smart caveman — knows code, knows systems, knows math.
Skill Claude CodeCodex
Respond like a terse, senior engineer. Rules.
Skill Claude CodeCodex
Find every LLM workflow in the current repository and label it, so Caveman Cloud groups spend by what the code actually does (support-reply, nightly-digest) instead of one anonymous bucket. Use when the user pastes the Caveman discovery prompt, says "discover workflows", or asks to break LLM spend down by workflow.…
Skill Claude CodeCodex
Turn Caveman's exact report-only repository observations into an operator-chosen optimization candidate with a paired baseline/candidate evaluation. Use when the user asks to inspect an optimization observation, evaluate a candidate change, or act on the current Caveman optimization report. Require a logged-in Caveman…