Caveman Saves Tokens by Doing Less, Not Just Saying Less | StuckInALocalMinima %F0%9F%9A%80"><br>Caveman promises an appealing optimization for coding agents:<br>Same answers, 65% fewer output tokens. Brain still big. Mouth small.
The idea is simple. Models often produce unnecessary explanations, narrate their tool calls, and pad conclusions with filler. Remove that verbosity and the agent should become faster and cheaper without becoming less capable.<br>Coding agents can consume a surprising number of tokens while inspecting repositories, running commands, and correcting their own mistakes. A small skill that removes wasted language sounds like an easy efficiency win. But a coding agent operates through a trajectory: call a tool, interpret the result, and decide whether to continue. An instruction that changes how the agent communicates can also change how it acts.<br>To measure that effect, I ran 60 tasks across Claude Code and Codex CLI using SWE Bench Verified, Terminal Bench 2, and USACO tasks. Caveman reduced Claude Code token usage by 18% and its reported cost by 25%, but in Codex token usage fell by only 3.9%.<br>The most important result was not how many tokens disappeared, but where they disappeared from. With Caveman enabled, Claude Code performed 17% fewer steps and made almost 20% fewer tool calls, and in Codex hidden reasoning tokens fell by 14%. Caveman didn’t merely make the agents speak less. It changed how much work they performed.<br>And if the skill mostly works by making the agent do less, adjusting the reasoning effort might be a better choice. Claude Code exposes reasoning effort as a native setting, and in this experiment medium effort reached roughly the same Pass@3 as Caveman with high effort while costing about 20% less. A style instruction that arrives at a similar operating point indirectly is harder to justify when a direct knob is one flag away.<br>What Caveman Promises<br>Caveman is a skill for Claude Code, Codex, Gemini, and other coding agents. It instructs the model to avoid narration, drop hedging, and communicate in compact fragments.<br>Its README summarizes the intended behavior with a memorable distinction:<br>Brain still big. Mouth small.
The main benchmark compares ten standalone answers, including prompts such as explaining Git rebase, React rerenders, and Docker multistage builds. The reported average falls from about 1,200 output tokens to under 300, the headline reduction of 65%.<br>That result is plausible for those prompts. They ask for prose, and Caveman aggressively compresses prose. The problem is the extrapolation.<br>A single conversational answer and a repository level coding task have very different token distributions. In a coding agent, useful output is source code, patches, and exact error messages. Caveman leaves much of that unchanged.<br>The project itself acknowledges this distinction. Its documentation warns that Caveman only targets output tokens, adds input tokens of its own, and may become net negative when the original model is already concise. It also describes the honest evaluation control as Caveman versus a simple terse instruction, rather than Caveman versus an unconstrained baseline.<br>Those caveats are good. Caveman is free, open source, and doesn’t collect telemetry. The issue is that the headline claims are much broader than the evidence supporting them.<br>The Experiment<br>I evaluated five agent configurations:<br>Configuration Agent Treatment claude-baseline Claude Code with Sonnet 5 High effort claude-caveman Claude Code with Sonnet 5 Caveman injected at session start (high effort) claude-medium Claude Code with Sonnet 5 Medium reasoning effort codex-baseline Codex CLI with GPT 5.6 Terra High effort codex-caveman Codex CLI with GPT 5.6 Terra Caveman forcibly activated (high effort) The benchmark contained 60 tasks: 20 from SWE Bench Verified, 30 from USACO, and 10 from Terminal Bench 2. Each task was executed three times under every configuration, for 900 trials in total.<br>The claude-medium configuration is particularly important. Comparing only baseline against Caveman mixes together two possible effects:<br>Caveman may compress communication.<br>Caveman may cause the model to spend less effort.<br>The medium effort arm provides a native way to reduce model effort without installing Caveman. It lets us ask whether the skill offers something beyond indirectly making the agent do less work.<br>Results<br>Claude Code showed substantial savings:<br>Metric Baseline Caveman Change Total tokens 1,595,199 1,302,271 18.4% lower Output tokens 19,860 16,059 19.1% lower Reported cost $1.003 $0.758 24.5% lower Steps 25.5 21.2 17.0% lower Tool calls 24.6 19.7 19.7% lower Agent prose characters 2,956 1,741 41.1% lower Pass@1 81.1% 80.0% 1.1 points lower Pass@3 91.7% 86.7% 5.0 points lower Codex showed a much smaller reduction:<br>Metric Baseline Caveman Change Total tokens 418,852 402,402 3.9% lower Output tokens 5,252 4,757 9.4% lower Reasoning tokens 2,542 2,200 13.5% lower Reported cost $0.257 $0.249 3.2%...