caveman-compress
juliusbrussee/caveman
Compress memory files into caveman-speak to reduce input tokens while keeping readable backups.
What is caveman-compress?
Compresses natural language files like CLAUDE.md or todo lists into abbreviated caveman format to save input tokens. The original is backed up out-of-tree and the compressed version overwrites the source. Use this when you want to shrink verbose documentation or preferences without losing the content.
- Detects file type and compresses natural language prose (markdown, text, typst) while preserving code blocks, URLs, commands, and technical terms exactly
- Removes articles, filler words, hedging, and redundant phrasing; uses short synonyms and sentence fragments
- Backs up the original as `<filename>.original.md` in an out-of-tree data directory so it won't be re-ingested by skill auto-loaders
- Validates compressed output and retries up to 2 times with targeted fixes if errors occur
- Skips code-only files (.py, .js, .json, .yaml, etc.) and never modifies code blocks or inline code
How to install caveman-compress
npx skills add https://github.com/juliusbrussee/caveman --skill caveman-compress- Python 3 installed
- The caveman skill installed via `npx skills add https://github.com/juliusbrussee/caveman --skill caveman-compress`
How to use caveman-compress
- 1.Trigger the skill with `/caveman-compress <filepath>` or ask the agent to compress a memory file
- 2.The skill detects the file type and calls Claude to compress it
- 3.The compressed version overwrites the original file
- 4.The human-readable backup is automatically saved to `$XDG_DATA_HOME/caveman-compress/backups/` (or Windows equivalent) with the `.original.md` suffix
- 5.If compression fails, the skill retries up to 2 times with targeted fixes; if still failing, the original file is left untouched and an error is reported
Use cases
- Compress a CLAUDE.md system prompt or memory file before each session to reduce token usage
- Shrink a verbose todo list or preferences file while keeping it readable
- Reduce the size of documentation or notes that get included in context windows
- Archive original versions of files before compression for reference without cluttering the working directory
- Developers using Claude Code or Cursor who manage large memory/context files
- Teams with verbose documentation that needs to fit within token budgets
- Anyone maintaining system prompts or preference files that grow over time
caveman-compress FAQ
Only natural language files: .md, .txt, .typ, .typst, .tex, and extensionless files. Code files (.py, .js, .json, .yaml, etc.) are never modified.
The original is backed up as `<filename>.original.md` in an out-of-tree data directory (`$XDG_DATA_HOME/caveman-compress/backups/<parent-dir-name>/` on Linux/Mac, `%LOCALAPPDATA%\caveman-compress\backups\<parent-dir-name>\` on Windows) so it won't be re-ingested by auto-loaders.
The skill retries up to 2 times with targeted fixes. If it still fails after 2 retries, an error is reported to you and the original file is left untouched.
Yes. Code blocks (fenced and indented), inline code, URLs, file paths, commands, technical terms, proper nouns, dates, and version numbers are copied exactly and never modified.
Yes. The original is always backed up as `<filename>.original.md` in the out-of-tree backup directory. You can restore it manually if needed.
Full instructions (SKILL.md)
Source of truth, from juliusbrussee/caveman.
name: caveman-compress description: > Compress a memory file such as CLAUDE.md or a todo list into caveman format to save input tokens, keeping a readable backup. Trigger: /caveman-compress.
Caveman Compress
Purpose
Compress natural language files (CLAUDE.md, todos, preferences) into caveman-speak to reduce input tokens. Compressed version overwrites original. Human-readable backup saved as <filename>.original.md, but NOT beside the source file — it lives in an out-of-tree data dir ($XDG_DATA_HOME/caveman-compress/backups/<parent-dir-name>/, or %LOCALAPPDATA%\caveman-compress\backups\<parent-dir-name>\ on Windows) so skill auto-loaders don't re-ingest it as a live file.
Trigger
/caveman-compress <filepath> or when user asks to compress a memory file.
Process
-
The compression scripts live in
scripts/(adjacent to this SKILL.md). If the path is not immediately available, search forscripts/__main__.pynext to this SKILL.md. -
From the directory containing this SKILL.md, run:
python3 -m scripts <absolute_filepath>
- The CLI will:
- detect file type (no tokens)
- call Claude to compress
- validate output (no tokens)
- if errors: cherry-pick fix with Claude (targeted fixes only, no recompression)
- retry up to 2 times
- if still failing after 2 retries: report error to user, leave original file untouched
- Return result to user
Compression Rules
Remove
- Articles: a, an, the
- Filler: just, really, basically, actually, simply, essentially, generally
- Pleasantries: "sure", "certainly", "of course", "happy to", "I'd recommend"
- Hedging: "it might be worth", "you could consider", "it would be good to"
- Redundant phrasing: "in order to" → "to", "make sure to" → "ensure", "the reason is because" → "because"
- Connective fluff: "however", "furthermore", "additionally", "in addition"
Preserve EXACTLY (never modify)
- Code blocks (fenced ``` and indented)
- Inline code (
backtick content) - URLs and links (full URLs, markdown links)
- File paths (
/src/components/...,./config.yaml) - Commands (
npm install,git commit,docker build) - Technical terms (library names, API names, protocols, algorithms)
- Proper nouns (project names, people, companies)
- Dates, version numbers, numeric values
- Environment variables (
$HOME,NODE_ENV)
Preserve Structure
- All markdown headings (keep exact heading text, compress body below)
- Bullet point hierarchy (keep nesting level)
- Numbered lists (keep numbering)
- Tables (compress cell text, keep structure)
- Frontmatter/YAML headers in markdown files
Compress
- Use short synonyms: "big" not "extensive", "fix" not "implement a solution for", "use" not "utilize"
- Fragments OK: "Run tests before commit" not "You should always run tests before committing"
- Drop "you should", "make sure to", "remember to" — just state the action
- Merge redundant bullets that say the same thing differently
- Keep one example where multiple examples show the same pattern
CRITICAL RULE:
Anything inside ... must be copied EXACTLY.
Do not:
- remove comments
- remove spacing
- reorder lines
- shorten commands
- simplify anything
Inline code (...) must be preserved EXACTLY.
Do not modify anything inside backticks.
If file contains code blocks:
- Treat code blocks as read-only regions
- Only compress text outside them
- Do not merge sections around code
Pattern
Original:
You should always make sure to run the test suite before pushing any changes to the main branch. This is important because it helps catch bugs early and prevents broken builds from being deployed to production.
Compressed:
Run tests before push to main. Catch bugs early, prevent broken prod deploys.
Original:
The application uses a microservices architecture with the following components. The API gateway handles all incoming requests and routes them to the appropriate service. The authentication service is responsible for managing user sessions and JWT tokens.
Compressed:
Microservices architecture. API gateway route all requests to services. Auth service manage user sessions + JWT tokens.
Boundaries
- ONLY compress natural language files (.md, .txt, .typ, .typst, .tex, extensionless)
- NEVER modify: .py, .js, .ts, .json, .yaml, .yml, .toml, .env, .lock, .css, .html, .xml, .sql, .sh
- If file has mixed content (prose + code), compress ONLY the prose sections
- If unsure whether something is code or prose, leave it unchanged
- Original file is backed up as FILE.original.md before overwriting — in the out-of-tree backup data dir (see Purpose), not beside the source file
- Never compress FILE.original.md (skip it)
Related skills
More from juliusbrussee/caveman and the wider catalog.
