Name: caveman-compress
Rating: 63.2 (1 reviews)
Author: JuliusBrussee

caveman-compress

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.

Quality

73%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

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

The canonical home for this skill is caveman-compress in JuliusBrussee/caveman

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 efficient, actionable, and notably strong on workflow clarity with validation and retry logic for a destructive operation. Minor conciseness tightening and slightly clearer reference structure would push it toward top marks.

Dimension	Reasoning	Score
Conciseness	The body is mostly lean — tight bulleted rules and before/after examples with no padding about concepts Claude already knows — though the CRITICAL RULE section partially restates the 'Preserve EXACTLY' code-block rule, leaving minor trimming opportunities at anchor 4 rather than 5.	4 / 5
Actionability	It provides a concrete executable command ('python3 -m scripts <absolute_filepath>'), specific compression rules, and copy-ready before/after examples, with only minor gaps (e.g., the exact CLI flag for the backup path is implied rather than shown), fitting anchor 4.	4 / 5
Workflow Clarity	The Process section gives a clear sequence with explicit validation checkpoints — detect, compress, validate, cherry-pick fix, retry up to 2 times, then report error and leave the original untouched — including a feedback loop for a destructive overwrite, matching anchor 5.	5 / 5
Progressive Disclosure	The body is well-organized into Purpose/Trigger/Process/Rules/Pattern/Boundaries sections and references real bundle files (scripts/__main__.py, cli.py, compress.py, detect.py, validate.py verified present) one level deep, but the bulk of behavior lives inline rather than split into separate reference files, sitting at anchor 4.	4 / 5
	Total	17 / 20 Passed

Description

66%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 clearly conveys what the skill does and when to invoke it, with a distinctive niche that avoids conflict risk. It could be strengthened with more specific action verbs and broader trigger-term coverage including synonyms and file extensions.

Suggestions

Add concrete action verbs beyond 'compress' (e.g., 'detect file type, compress prose, preserve code blocks, validate output') to lift specificity toward anchor 5.

Broaden trigger terms with natural synonyms users might say (e.g., 'shrink CLAUDE.md', 'reduce token usage', 'minify memory file') and file extensions (.md, .txt) for better trigger coverage.

Expand the 'when' clause to list multiple trigger phrases rather than a single slash command to strengthen completeness.

Dimension	Reasoning	Score
Specificity	Names the domain ('memory file such as CLAUDE.md or a todo list') and concrete actions ('Compress ... into caveman format', 'save input tokens', 'keeping a readable backup'), but the actions stay fairly generic and coverage is not comprehensive, fitting anchor 3 better than 4.	3 / 5
Completeness	Both 'what' (compress memory files into caveman format, keep a backup) and 'when' ('Trigger: /caveman-compress') are explicit; the trigger guidance is concrete but somewhat narrow, so it sits at anchor 4 rather than 5.	4 / 5
Trigger Term Quality	Relevant keywords are present ('memory file', 'CLAUDE.md', 'todo list', 'compress') but it lacks common synonyms or file-extension variants users might say, landing at anchor 3 rather than 4.	3 / 5
Distinctiveness Conflict Risk	The 'caveman format' niche is highly distinctive with a dedicated /caveman-compress trigger, giving minimal overlap risk with other skills — a clear anchor 5 match.	5 / 5
	Total	15 / 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: plugins/caveman/skills/caveman-compress/SKILL.md
Commit: df2ccd8

Reviewed: 4 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.