A response-style skill that compresses writing into short, telegraphic fragments while preserving code, commands, paths, identifiers, URLs, numbers, and error messages exactly.

Producing terse responses, compressed explanations, and byte-exact technical output when the user enables this mode.

It reduces filler when a task calls for very concise output without changing technical text that must remain exact.

caveman — Skill for Claude Code by Bumblebiber

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent

npx skills add Bumblebiber/its-over-9k --skill caveman

Clone the repo

git clone --depth 1 https://github.com/Bumblebiber/its-over-9k

Made for: Claude Code.

Or install o9k-caveman, the plugin that ships this one along with the rest of its 1 skill.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/bumblebiber/its-over-9k/caveman.svg)](https://agentmods.dev/skills/bumblebiber/its-over-9k/caveman)

Your own site

<a href="https://agentmods.dev/skills/bumblebiber/its-over-9k/caveman"><img src="https://agentmods.dev/badge/skills/bumblebiber/its-over-9k/caveman.svg" alt="Measured on agentmods" height="20"></a>

Per session 66 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 610 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.00066	$0.00610
Opus 5	$0.00033	$0.00305
Sonnet 5	$0.00013	$0.00122
Haiku 4.5	$0.00007	$0.00061

Measured 6d ago against content hash d759672790b0, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

caveman scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugins/o9k-caveman/skills/caveman/SKILL.md · 65 lines

How it starts

The opening of the file, as written. The whole thing — 65 lines — stays where its author put it; the contents beside it link to each section on GitHub.

caveman — Output Compression

Compress every response. Information survives; filler dies. Adapted from JuliusBrussee/caveman (MIT).

What dies

Pleasantries, hedging, apologies ("I'd be happy to", "It seems that", "Great question")
Restating the request or your previous answer
Articles and connective tissue where meaning survives without them
Transitions ("Now let's", "Next, we will"), sign-offs, offers of further help
Explanations nobody asked for

What survives byte-exact — NEVER compress

Code, commands, file paths, identifiers, URLs
Error messages and log lines (verbatim — they get grepped)
Numbers, versions, IDs

The pattern

[thing] [state/action] [reason]. [next step].

Before (41 tokens):

I looked into the issue and it seems that the authentication middleware is rejecting the token because the clock skew between the two servers exceeds the allowed tolerance. We should probably increase the leeway setting.

After (17 tokens):

Auth middleware rejects token — clock skew between servers exceeds tolerance. Fix: raise leeway setting.

Levels

Level	Rule	When
`lite`	Drop filler + hedging, keep full sentences	Default for user-facing explanation
`full`	Telegraphic fragments, pattern above	Default for status updates, findings, tool narration
`ultra`	Keywords + code only, no sentences	User opted in explicitly; agent-to-agent traffic

User says "caveman lite/full/ultra" → switch level. "caveman off" → normal prose.

Auto-revert — compression never outranks safety

Use complete, explicit prose regardless of level for:

Security warnings and vulnerability reports
Destructive or irreversible actions (deletes, force-pushes, migrations, spends)
Ambiguous multi-step instructions the user must execute exactly
Content that leaves the conversation: commit messages, PR bodies, docs, emails
Legal, compliance, or safety-relevant statements

Read the full file on GitHub · 65 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

6d ago First seen · 65 lines · 66 tokens per session scan A d759672790b0

Subscribe to this mod's changes

caveman is a skill published in the GitHub repository Bumblebiber/its-over-9k (0 stars, last pushed 9d ago), licensed MIT. It adds 66 tokens to every session and 610 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

local-ai-agents

Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…

microsoft/ai-agents-for-beginners · 200 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

next-partial-prefetching-adoption

Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…

vercel/next.js · 103 tokens

chronicle

Analyze Copilot session history for standup reports, usage tips, session search, and session reindexing. Use when the user asks for a standup, daily summary, usage tips, workflow recommendations, wants to search or find past sessions by keyword/file/PR, wants to reindex their session store, or asks about deleting…

microsoft/vscode · 72 tokens

babysit-pr

Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…

openai/codex · 114 tokens