#1Caveman
Why use many token when few token do trick. Viral skill + proxy for coding agents that cuts 65% of tokens by talking like a caveman.
$
/plugin marketplace add JuliusBrussee/cavemanBest of 2026 · updated 2026-10-15
Cut Claude Code token usage
Claude Code plugins and skills whose own description focuses on tokens, context-window size, compression or cost, ranked by GitHub stars. We list what they claim; check each README for benchmarks.
143 matches · Ranked by real GitHub stars. Sponsored slots are labelled and never change the ranking.
| # | Name | Stars |
|---|---|---|
| 1 | Caveman JuliusBrusseeGo · Apache-2.0 · 18h ago | 111,953 |
| 2 | Claude Mem thedotmackTypeScript · Apache-2.0 · 1d ago | 100,811 |
| 3 | rtk rtk-aiRust · Apache-2.0 · 12h ago | 83,395 |
| 4 | 9router decoluaJavaScript · MIT · 1w ago | 30,821 |
| 5 | Img2threejs img2threejsPython · Apache-2.0 · 1w ago | 18,156 |
| 6 | Ego Lite citrolabsJavaScript · MIT · 1d ago | 17,403 |
| 7 | Prompt Master nidhinjsMIT · 3d ago | 14,467 |
| 8 | MemOS MemTensorTypeScript · Apache-2.0 · 5d ago | 11,869 |
| 9 | Codeburn getagentsealTypeScript · MIT · 22h ago | 11,449 |
| 10 | Tokscale junhoyeoRust · MIT · 9h ago | 5,677 |
| 11 | Router weave-osGo · Apache-2.0 · 14h ago | 5,593 |
| 12 | Axonhub loopljGo · Apache-2.0 · 9h ago | 5,358 |
| 13 | Cost Xray tigerless-labsPython · MIT · 1d ago | 4,968 |
| 14 | Claude Devtools matt1398TypeScript · MIT · 2w ago | 3,976 |
| 15 | CPA Manager Plus seakeeGo · MIT · 9h ago | 3,845 |
Persistent Context Across Sessions for Every Agent – Captures everything your agent does during sessions, compresses it with AI, and injects relevant context back into future…
$/plugin marketplace add thedotmack/claude-mem- #3rtk
CLI proxy that reduces LLM token consumption by 60-90% on common dev commands. Single Rust binary, zero dependencies
$brew install rtk - #49router
Unlimited FREE AI coding. Connect Claude Code, Codex, Cursor, Cline, Copilot, Antigravity to FREE Claude/GPT/Gemini via 40+ providers. Auto-fallback, RTK -40% tokens, never hit…
$npm install -g 9router Rebuild the object in a reference image as a code-only, procedural, quality-gated, animation-ready Three.js model. Token-efficient image-to-3D.
$git clone https://github.com/img2threejs/img2threejs- #6Ego Lite
The fastest browser for AI agents to run browser automation, built for sharing your logged-in browser state with your AI agents, like Codex or Claude Code, without disturbing…
$npx skills add citrolabs/ego-lite A Claude skill that writes the accurate prompts for any AI tool. Zero tokens or credits wasted. Full context and memory retention
$git clone https://github.com/nidhinjs/prompt-master- #8MemOS
Self-evolving memory OS for LLM & AI Agents: ultra-persistent memory, hybrid-retrieval, and cross-task skill reuse, with 35.24% token savings and DeepSeek Harness support.
$npx @deepseek-ai/dsh plugin --profile web add @memtensor/memos-cloud-dsh-plugin@latest - #9Codeburn
Free, local tool to track AI coding token usage and cost across 37 tools and agents (Claude Code, Cursor, Codex, Gemini and more), by model, project, and task. npx codeburn
$claude mcp add codeburn -- npx -y codeburn mcp - #10Tokscale
Track token usage across AI coding agents from your terminal. Global leaderboard with trillions of tokens tracked.
$npx tokscale@latest - #11Router
Model router for agentic systems. Routes every prompt to the right model in <50ms. Cut costs 40-70% with just an endpoint change.
$npx @weave-os/router - #12Axonhub
Open-source AI Gateway — Use any SDK to call 100+ LLMs. Built-in failover, load balancing, cost control & end-to-end tracing.
$git clone https://github.com/looplj/axonhub - #13Cost Xray
See what Claude Code and Codex actually send to the API — and what each part costs.
$git clone https://github.com/tigerless-labs/cost-xray The missing DevTools for Claude Code — inspect session logs, tool calls, token usage, subagents, and context window in a visual UI. Free, open source.
$brew install --cask claude-devtoolsA self-hosted CPA / CLIProxyAPI management panel and AI gateway observability dashboard for requests, usage, cost, quota, failures, and account health.
$git clone https://github.com/seakee/CPA-Manager-Plus- #16Abtop
Like htop, but for AI coding agents. Monitor Claude Code & Codex CLI sessions, tokens, context window, rate limits, and ports in real-time.
$cargo install abtop - #17Vibe Skills
Intelligent Skill routing and workflow orchestration for AI agents — +21.12 pp reward, −29.6% tokens on SkillsBench with DeepSeekV4Flash-VE.
$git clone https://github.com/foryourhealth111-pixel/Vibe-Skills Local-first desktop widget for tracking token usage, costs, limits, and session activity across 46+ AI coding tools—including Claude Code, Codex, Cursor, OpenCode, OpenClaw,…
$npx wrangler loginFind the ghost tokens. Fix them. Survive compaction. Avoid context quality decay.
$/plugin marketplace add alexgreensh/token-optimizer- #20Tweakcc
Customize Claude Code's system prompts, create custom toolsets, input pattern highlighters, themes/thinking verbs/spinners, customize input box & user message styling, support…
$npx tweakcc - #21Openwolf
Portable project memory across Claude Code, Codex and OpenCode, plus token accounting measured from harness transcripts. Local file I/O, no API calls, no telemetry.
$npm install -g openwolf - #22Claude Usage
A local dashboard for tracking your Claude Code token usage, costs, and session history. Pro and Max subscribers get a progress bar. This gives you the full picture.
$brew install phuryn/claude-usage/claude-usage - #23TokenTracker
Local-first AI token usage & cost tracker for 31 coding tools incl. Claude Code, Codex, Cursor, Gemini & DeepSeek Harness—with native apps. Never reads prompts.
$npx tokentracker-cli - #24Design Dna
Turn reference UIs (images, screenshots, URLs) into quantified Design DNA JSON—tokens, qualitative style, visual effects—then generate matching UI from your content.
$npx skills add zanwei/design-dna Turn Claude into a senior design architect: DTCG tokens, 52 components, WCAG 2.2 AA-AAA, 138 design systems, any-framework code, and 50 objective gates that fail the build…
$claude mcp add ux-ui-gates -- npx -y --package=ux-ui-agent-skills ux-ui-mcpGhostty Blackhole puts a real, ray-traced black hole inside your terminal. It grows as Claude Code's context window fills up, live. A fresh session is a quiet hole in the…
$git clone https://github.com/s0xDk/ghostty-blackholeTopic → 4K narrated video for coding agents. v5.3.0: local TTS (edge free + azure, no external engine), manifest-based Asset Engine, Remotion composition, cost-gated AI…
$npx skills add Agents365-ai/video-podcast-maker/skills/video-podcast-maker -gMake your agents talk human. ADHD-friendly output styles for Claude Code, Codex, and others. So you can pay attention, not tokens.
Agent skill★ 1,450$/plugin marketplace add alexgreensh/attention-spanAn IOS Simulator Skill for Agents. Use it to optimise their ability to build, run and interact with your apps, and to proxy xcodebuild to save token and context wastage.
$/plugin marketplace add conorluddy/ios-simulator-skill- #30Styleseed
Open-source design-method engine for Claude Code, Codex & Cursor. 23 agent skills for fixed design judgment, multiple grammars, semantic palettes, reference compilation, and…
$npx -y skills add bitjaru/styleseed -a codex claude-code -y --copy
