跳到主要内容
Supermarket
返回能力市场
Agent Pack
programming
NOASSERTION

caveman

🪨 why use many token when few token do trick — Claude Code skill that cuts 65% of tokens by talking like caveman

JuliusBrusseeJuliusBrussee
59/ 100

公开评测 · 综合采用结论

存在需要人工复核的风险或证据不足

查看评测依据 查找我项目的报告基于公开项目证据,非安全认证或安装推荐
105.6kstars
6.1kforks
最近更新 2个月前
评测生成时间(北京时间)
本报告引擎
v3.0.0
当前引擎
v3.17.3

本报告与当前引擎使用不同规则;原分数不会自动更新,不同版本的分数不宜直接对比。

重新评测此项目

进入后确认来源与额度,提交才会创建任务。

代码版本未记录或记录无效,无法核对本报告的 commit 与送评文本;不会用当前项目代码回填历史记录。

Evaluation report

综合采用结论

59
D
满分 100 · 静态采用参考
谨慎采用高风险
决策摘要

存在需要人工复核的风险或证据不足

85%
高置信度
100
文档
78
安全
95
质量
100
活跃
70
采用

历史报告未保存置信度明细;重新评测后可查看各项证据贡献。

本报告评了什么

综合分用于静态采用筛选,不是任务成功率,也不代表使用 Skill 后的能力提升。

静态检查范围
记录 25 个文件 · 104,326 字符
只检查采集到的受限文本与已知风险模式;未读取的脚本、依赖和运行环境不在结论内。
AI 内容复核
未使用
复核送评文本的说明和设计,不是安装测试或实际任务执行。记录缺失时不推断已使用或未使用。
真实任务效果
本报告未提供实测证据
没有使用/不使用 Skill 的任务对照、重复运行或耗时与 token 成本结果。采用前仍需在隔离环境验证自己的任务。

脚本引用覆盖

脚本引用覆盖记录不可得;不推断没有引用或已经检查。

How it works · 未生成图示

本次报告未完成图示提取

本次评测未完成 AI 证据提取;确定性评分与安全扫描仍然有效。重新评测后,证据充分时会自动选择合适图型。

补充参与方或组件说明谁参与、各自负责什么
说明关系与顺序提供输入输出、调用或依赖证据
重新评测自动选图按证据选择流程、时序或架构图
五维表现
确定性工程质量 95/100
采用建议
优势
  • 问题与用途描述
  • 有效 README
  • 安装或接入步骤
  • 可执行示例
关注点
  • 发现高风险的一键下载执行或安装命令

也有自己的公开项目?先免登录查已有报告;没有报告时,可保留地址继续登录评测。

查找我项目的报告 →
文档证据
100/100
问题与用途描述10 分
有效 README12 分
安装或接入步骤14 分
可执行示例16 分
输入、参数或工具说明11 分
输出或结果说明9 分
限制、权限或边界12 分
错误处理或排障8 分
许可证信息5 分
结构化章节3 分
安全证据
高风险
发现高风险的一键下载执行或安装命令
unsafe-install-commandREADME.md:87high confidence
curl -fsSL https://raw.githubusercontent.com/JuliusBrussee/caveman/v2.2.0/install.sh | bash

修复:固定版本与校验和,先下载审查再执行,避免管道直接交给 Shell。

优先改进清单
  1. 01固定版本与校验和,先下载审查再执行,避免管道直接交给 Shell。
方法、证据与局限展开
数据来源

GitHub Repository API

扫描范围

25 个文件 · 104,326 字符

评测引擎

v3.0.0 · AI 内容复核:未使用

局限
  • 静态评测不会安装或执行项目代码
  • 安全扫描基于高信号文件与已知模式,不能替代人工审计
  • 流行度只反映采用程度,不代表安全或工程质量

README

Caveman

why many token when few do trick

Caveman make your AI agent say less and read less. Code stay exact. Brain still big.

GitHub stars npm downloads 30+ agents License

33.2% fewer

input tokens, whole sessions
54 Claude Code runs, 18 of 18 answers right

54% cheaper

with Auto, same score
85 coding tasks vs Opus 5.5 at xhigh, interim

397 → 76

tokens for one test run
Blocks vs | tail -20

1.4 to 2.4× cheaper

with caveman-style output
Adobe Research, eight models

Cited by Adobe Research · A/B tested by JetBrains · Remade for Elasticsearch on Elasticsearch Labs
#1 on Hacker News · #1 on GitHub Trending · "No way this actually works." ThePrimeagen

Install · Six rocks · Proof · Say less · Read less · Auto · Blocks · Commands · Caveman Cloud


Normal agent · 63 tokens Caveman agent · 20 tokens

The reason your React component is re-rendering is likely because you're creating a new object reference on each render cycle. When you pass an inline object as a prop, React's shallow comparison sees it as a different object every time, which triggers a re-render. I'd recommend using useMemo to memoize the object.

New object ref each render, so React re-renders. Wrap the prop in useMemo.

Same fix. 63 token become 20. Brain still big.

Pick your club:

SkillSame answerTokens
/cavemanNew object ref each render, so React re-renders. Wrap the prop in useMemo.20
/ultracaveInline object prop, new ref, re-render. useMemo.14
/megacave新參照致重繪。useMemo。13

Token counts: tiktoken o200k.

Install

npx @caveman-ai/cli@latest

One command. Mac, Linux, Windows. Caveman find your agents and your logins, show what it change. Press Enter. Done. Needs Node.js 22.13+.

After that, plain claude stays caveman'd. So do Codex, Gemini CLI, OpenCode, and 30+ more. It runs on your machine, with your keys and your Claude Pro/Max login. One rock. That it.

Had caveman before 4.0? Say @latest, or the old copy on your machine wins. Already set up? caveman setup opens the screen again. Only want shorter answers? Get just the skill.

Six rocks, one install

New in 4.0: caveman is six parts. All on after install. Each has its own switch.

RockWhat it doesMeasured
Say lessAgent answers short. Code stays exact63 → 20 tokens, same fix
Read lessLogs, JSON, CSV, test output shrink before the model reads them33.2% fewer input tokens per session
Auto · newPicks the model and the effort for each ask54% cheaper than Opus at xhigh, same score
Blocks · newCode mode: agent keeps the scripts it writes and runs them again397 → 76 tokens for one test run
Waste fixesFinds where your tokens go and fixes it. New: keeps the prompt cache warm while you pauseNot measured yet
BrowseWeb pages come back compressed129.8× smaller page

caveman status shows what is on. caveman off browse turns one off. Only Auto needs an account, and it is free. 4.0.0 release notes

Proof

Outside labs first. Ours come with each rock below. Nothing rounded up. Red rows stay red.

WhoSetupResult
Adobe Research, CAVEWOMAN paper (cites this repo)Caveman-style output, eight models, five datasetsCost cut 1.4 to 2.4× per model, up to 3×
Elastic, Elasticsearch LabsCaveman mode remade for Elasticsearch, eight live MCP scenarios63.6% fewer response tokens. "Zero information loss."
JetBrains86 real coding tasks, paired A/BNo measurable quality loss (p = 0.82). 8.5% fewer output tokens

Two findings shaped caveman. Adobe: cavemanning your prompt makes answers worse, so caveman never touches your prompt. JetBrains: agent sessions are mostly code and tool calls, which the voice leaves alone. So caveman grew rocks that shrink what the agent reads and what it runs on.

Say less: the caveman voice

A voice, not broken grammar. Every reply follows the same structure.

RuleWhat it means
Answer first[thing] [action] [reason]. [next step]. No greeting, no "let me", no recap
One idea per sentenceBuilt on ASD-STE100, the controlled English of aircraft maintenance manuals: 20 words max, active voice, one term per thing
Meaning never droppedArticles can go. not, never, no, only never go. Numbers and units stay exact
Payload verbatimCode, commands, paths, and error messages untouched, character for character
No theaterNo "me think". No chatter between tool calls. If caveman phrasing isn't shorter, plain wins
Knows when to stopSecurity warnings, irreversible actions, and confused users get full sentences. Then grunt resumes
Your prompts stay yoursNever rewritten. Research say that backfire

Ten dev questions on claude-opus-5-5:

InstructionOutput tokens
None6,983
Answer concisely.4,334
/caveman4,119
/ultracave2,693

"Be concise" is the real baseline; new models already know it. On top of it, /caveman cuts 3% more at the median and /ultracave 35% more. Measured on the 3.1.0 rules, not yet re-run. The rules cost about 1,160 tokens of input, and if you pay per request (GitHub Copilot), shorter answers save nothing. Harness · Every case where caveman loses

Only want this rock? The skill alone, no proxy:

npx skills add JuliusBrussee/caveman -g

skills.sh

Type /caveman ($caveman in Codex) if it doesn't start on its own. Say stop caveman to go back.

Other ways in: Claude app and phone, Claude Code plugin, Gemini, every agent at once, Windows, uninstall

Claude on the web, the phone app, or Cowork? No terminal. Add caveman to your Claude account once, in Customize > Plugins: steps. Not tested end to end yet.

# Claude Code plugin, auto-starts every interactive session, subagents too
claude plugin marketplace add JuliusBrussee/caveman && claude plugin install caveman@caveman

# Gemini CLI
gemini extensions install https://github.com/JuliusBrussee/caveman

# Every agent on your machine at once, plus the Claude Code statusline badge (Node.js 22.13+)
curl -fsSL https://raw.githubusercontent.com/JuliusBrussee/caveman/v4.0.1/install.sh | bash

Already inside Claude Code? Same rock, slash form. Type these where you talk to Claude:

/plugin marketplace add JuliusBrussee/caveman
/plugin install caveman@caveman

Windows, PowerShell 5.1+:

irm https://raw.githubusercontent.com/JuliusBrussee/caveman/v4.0.1/install.ps1 | iex

On npm 12 or newer, new npm block git install. One-liners above handle it themselves from v3.2.0 on. Old one-liner pinned to v3.1.0 or older still fail there: use one above, or this: npx --allow-git=root -y github:JuliusBrussee/caveman#v4.0.1

Changed your mind: npx -y github:JuliusBrussee/caveman -- --uninstall (on npm 12+, add --allow-git=root too)

Only grabbed the skill? npx skills remove JuliusBrussee/caveman -g take it back out.

Install broke? Open your agent in this repo and say "Read CLAUDE.md and INSTALL.md, install caveman for me." Agent fix own brain.

Read less: 33.2% fewer input tokens

The voice shrinks what the agent says. This shrinks what it reads. Same answers: 18 of 18 right.

File typeThe file, through cavemanFile savedWhole Claude Code session, 3 runsSession saved
CSV28,041 → 31498.9%165,823 → 74,48455.1%
Logs22,810 → 34898.5%148,807 → 74,06850.2%
YAML20,447 → 17899.1%132,124 → 71,02746.2%
Test output18,806 → 20398.9%150,377 → 108,51427.8%
JSON18,837 → 28198.5%147,975 → 108,93926.4%
HTML21,670 → 21,670none yet140,687 → 154,6419.9% worse
All six130,611 → 22,99482.4%885,793 → 591,67333.2%

The file is each test file through today's compressor alone. The session is the whole agent run, which also carries the system prompt and tools, so it moves less. Honest part: the session run is from August 2026 on an older engine, HTML has no compressor yet so it lost, and each test file hides one record in 61 to 72 KB of noise. Yours will vary. Headroom on the same suite: 15 of 18 right, 6.7% fewer tokens. Method

Every original stays on your disk. The agent can pull it back any time.

Auto: 54% cheaper, same score

New model in your agent's picker. Pick Auto and caveman chooses the model and the effort for each ask. Hard ask, big brain. Easy ask, save rocks.

SetupTasks solvedCost per tryMedian time
Opus 5.5, xhigh effort, every ask44.3%$7.8335.2 min
Auto, Claude models only43.7%$3.5915.9 min
Auto, plus OpenAI and open models42.9%$3.4716.7 min

Auto costs 0.46× what Opus 5.5 at xhigh costs. That is 54% saved, in under half the time. 85 real coding tasks from SWE-Interact and SWE-Together, one try each. Honest part: the run is half done, the 0.5-point score gap is too small to tell apart at this size (±7), and most of the saving is Auto lowering effort on easy asks. Already on Opus at medium? Expect less; not measured yet. Full table and limits

caveman login                  # free account. Auto shows up in the model picker
caveman providers add openai   # optional: give Auto more models, on your own key

Works in Claude Code, OpenCode, and Codex on a ChatGPT login, with API keys, Claude Pro/Max, or ChatGPT. Any other model you pick runs as picked. Not signed in: no Auto, nothing sent. What Auto sends

Blocks: code mode, script written once

skills.sh for code mode. Code mode is the agent writing a script instead of making tool call after tool call (Cloudflare, Anthropic). Problem: agent write same script every session, throw it away, write it again. Blocks keep it in your repo, check it every pull request, hand it back.

InputRawWhat agents type nowBlock
go test -v ./..., 326 tests10,561397 · | tail -2076 · test-summary
1.4 MB JSON file251,893236 · their own heredoc227 · json-peek
366-line Go file3,111250 · grep -n489 · grep-defs

Tokens the agent reads back. In one week, agents on one machine wrote the same script 394 times. Purdue measured the habit at up to 9.57% of task cost. grep-defs loses to a bare grep; row stays. Method

Comes with caveman. Eight blocks ready: json-peek · jsonl-stats · test-summary · first-error · wait-for · grep-defs · http-json · replace-in-file. When the agent starts typing a script a block already covers, it gets one line: use the block. Never denies or rewrites a command.

caveman-blocks scan    # what your agents retyped this month. Script text never leaves your machine
caveman-blocks init    # rules and block index into this repo's AGENTS.md

Pre-release (v0.1.0-rc.5). Python only, no sandbox yet. Own repo, own stars: caveman-ai/blocks

Smaller rocks

RockWhat it doesMeasured
caveman learnRanks where your tokens go, from agent history already on disk. caveman learn implement applies fixes one diff at a time, only on your yesNot yet
Warm cache · newStep away and the provider forgets your session, so the next ask pays full price. Caveman keeps Anthropic's prompt cache warm while you pause, only when your own history says you come back in time. Every warm is listed in caveman statsNot yet
caveman browse <url>A compressed web page for the agent121 tokens instead of 15,704 for a Playwright snapshot of a 200-row table. Tiny forms lose 2.3×. Bench
/caveman-compress <file>Shrinks memory files like CLAUDE.md, keeps a backup4,138 tokens instead of 6,198 across five files, 22.8% to 49.1% each, every heading, code block, and path intact. Bench
MiddlewareSame shrinking inside the Vercel AI SDK, LangChain, OpenAI, or Anthropic call you already make. TypeScript · Python · Every framework · One container for the teamNot yet

Caveman Learn report: a summary and savings cards on the left; the biggest places tokens go, with one fix opened, and a chart of how full sessions get on the right

Commands

In your terminal:

caveman claude                # start an agent on caveman: codex · gemini · opencode · aider · kilo · qwen · hermes · openclaw · pi
caveman status                # what is on, which agents it reaches, one next step
caveman on browse             # switch one rock on or off: output · input · waste-fixes · routing · scripts · browse
caveman login                 # free account, turns on Auto
caveman learn                 # where your tokens go
caveman trial -- claude       # A/B a real session on your own work
caveman shrink -- pnpm test   # compress noisy command output
caveman convert --dry-run     # pixel mode: skills the model reads as images
caveman stats                 # your token history
caveman doctor                # one problem per line, with its fix

In your agent:

CommandWhat it does
/caveman · /ultracave · /megacaveThe voice, the grunt, the 文言文. /caveman status shows the mode, /caveman off stops it
/caveman-commitOne-line Conventional Commit
/caveman-reviewOne finding per line: L42: 🔴 null deref. Guard it.
/caveman-compress <file>Shrinks memory files and backs up the original
/caveman-statsReal token usage for this Claude Code session
/caveman-helpEvery mode and command on one screen
cavecrewSubagents that find, edit, and review code, then report back in caveman
investigate-first · lean-build · surgical-patch · safe-refactor · migration · verify-and-stopWork patterns that write less code. Your agent picks them up when a task fits
caveman-setup · caveman-discover · caveman-learn · caveman-manage · caveman-optimize · caveman-explore · caveman-evidence-reviewDrive the proxy from inside your agent

Caveman Cloud

Everything above is for the coding agent you use. Caveman Cloud is for the agents you build.

You ship an agent. Every run costs money. Is it good? Is it cheap? Which change made it worse? Cloud answers.

Cloud doesWhat you get
ObservabilityEvery run traced to the work and what it cost. Sets itself up from your repo, your keys, and the logs or OpenTelemetry you already have
EvalsGrades how good the answers are, on your own tasks, from your real traffic
AutopilotProposes a cheaper change, tests it on your recorded work, ships only when the evals say answers are still as good. You approve the pull request
ProofEvery number says if it is measured, inferred, or verified. A saving counts only after real traffic shows it

Sign up free  ·  Try the demo  ·  Book a call  ·  Docs

In the wild

ThePrimeagen reacts to Caveman: No way this actually works

ThePrimeagen: "No way this actually works." Watch

Adobe Research: CAVEWOMAN, arXiv, June 2026

JetBrains: "It is fun, and it costs you nothing measurable in quality." Read

Elastic: elastic-caveman on Elasticsearch Labs, April 2026

Hacker News: #1, 904 points

GitHub Trending #1 overall, July 2026 · Trendshift #1 repo of the day, April 2026 · Product Hunt #8 of the day

Started as a joke in April 2026. Now past 100,000 stars. Joke got serious. Voice did not.


Star History Chart

Privacy

  • The skill runs on your machine and sends nothing. One exception: /caveman-compress sends the file you compress to your configured model provider.
  • The CLI sends usage stats by default: commands run, token counts, caveman-blocks commands and usage counts, a random install ID, OS, and IP. Never your prompts, code, file paths, or scripts you wrote. One person maintains this for free, and those stats show what to build next. Off for good, Blocks included: caveman telemetry off or DO_NOT_TRACK=1.
  • Auto, only once you sign in and pick it: each ask sends Caveman Cloud your latest ask (with whatever your agent attaches to it, such as CLAUDE.md, memory or @-mentioned files), the one before it, and the end of the agent's last reply, plus each request's tool names, effort, a few agent headers, and the previous request's token counts. On the Free plan Caveman may keep it to improve routing. caveman off routing stops it.
  • Where requests go: from your machine to your provider, or to another provider you added with caveman providers add, on your own key. Only a request routed to a Caveman Cloud model passes through Caveman Cloud, whole (system prompt, tool results, code), on your Caveman project's stored key, billed to that project. caveman providers cloud off stops that.

The full list, and how to delete sent telemetry: SECURITY.md.

License

Apache-2.0, whole repo. Read it, fork it, ship it, host it. Free like mammoth on open plain. Older releases and third-party notices: LICENSING.md. "Caveman" and the rock logo are trademarks of Julius Brussee.

Cite

@software{brussee2026caveman,
  author = {Brussee, Julius},
  title  = {Caveman: why many token when few do trick},
  year   = {2026},
  url    = {https://github.com/JuliusBrussee/caveman}
}

🪨 Caveman save you token. Star cost zero. Fair trade.

Docs · Install matrix · Honest numbers · Contributing · Caveman Cloud · Issues