caveman-compress

Compress natural language memory files (CLAUDE.md, todos, preferences) into caveman format to save input tokens. Preserves all technical substance, code, URLs, and structure. Compressed version overwrites the original file. Human-readable backup saved as FILE.original.md. Trigger: /caveman-compress FILEPATH or "compress memory file"

Quality

78%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

High

Do not use without reviewing

Fix and improve this skill with Tessl

tessl review fix ./skills/caveman-compress/SKILL.md

Quality

Content

81%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The skill body is well-structured and actionable, with a clear multi-step workflow that includes validation, retry, and original-restoration feedback loops appropriate to its destructive overwrite operation. Its main weaknesses are minor verbosity in the backup-path rationale and inlined rule lists that could be trimmed or externalized.

Dimension	Reasoning	Score
Conciseness	The body is well-organized and mostly lean, assuming Claude's competence and listing rules tersely; minor verbosity appears in the Purpose/backup-path detail and the duplicated Preserve-EXACTLY/code-block rules.	4 / 5
Actionability	It gives a concrete executable invocation ('python3 -m scripts <absolute_filepath>') and a numbered CLI behavior flow, plus concrete before/after compression examples; minor gaps are that it does not show how to run from a path other than the SKILL.md directory.	4 / 5
Workflow Clarity	The destructive overwrite workflow has an explicit validate -> fix -> retry (up to 2) -> restore-original-on-failure loop, and the scripts implement backup-before-overwrite plus validation checkpoints, satisfying the feedback-loop requirement for batch/destructive operations.	5 / 5
Progressive Disclosure	The SKILL.md is a concise overview that offloads executable logic to scripts/ (verified present: __main__.py, cli.py, compress.py, detect.py, validate.py), with a one-level-deep reference to scripts/__main__.py; it is slightly below 5 only because the compression rules are inlined rather than referenced.	4 / 5
	Total	17 / 20 Passed

Description

76%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is mostly strong: it clearly states what the skill does and gives explicit trigger guidance, and the caveman-compression niche is distinctive. Its main weakness is trigger term coverage, which leans on one slash-command and one phrase rather than the natural vocabulary users might invoke.

Suggestions

Add natural-language synonyms users might say, e.g. 'compress', 'shrink', 'reduce tokens', 'summarize', or 'condense memory/CLAUDE.md files' alongside the existing trigger phrase.

Mention the supported file types or extensions (e.g. '.md, .txt, extensionless memory files') so the trigger surfaces on file-extension cues as well as phrasing.

Dimension	Reasoning	Score
Specificity	It lists several concrete actions ('Compress', 'overwrites the original file', 'saves backup', 'Preserves all technical substance, code, URLs, and structure') covering the core behavior with only minor gaps (no mention of validation or retry).	4 / 5
Completeness	It explicitly answers both 'what' (compresses natural language memory files into caveman format, preserves technical substance) and 'when' ('Trigger: /caveman-compress FILEPATH or "compress memory file"').	5 / 5
Trigger Term Quality	It supplies the trigger phrases '/caveman-compress FILEPATH or "compress memory file"' and mentions 'memory files (CLAUDE.md, todos, preferences)', but lacks common synonyms a user might naturally say (e.g. 'shrink', 'reduce tokens', 'summarize', '.md').	3 / 5
Distinctiveness Conflict Risk	'Caveman format' compression of memory files is a clear niche with a distinct trigger; minor overlap risk exists only with generic summarization or compression skills.	4 / 5
	Total	16 / 20 Passed

Validation

100%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: JuliusBrussee/caveman
Path: skills/caveman-compress/SKILL.md
Commit: 14d4f2e

Reviewed: 5 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.