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

84%

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

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.

A well-structured, actionable body with a clear validated workflow for a destructive operation and good organization of its script bundle. Slight conciseness redundancy and limited use of separate reference files keep the lower-weight dimensions at 4.

Suggestions

De-duplicate the code-block preservation guidance: state it once (e.g., in 'Preserve EXACTLY') and reference it from the CRITICAL RULE instead of repeating the full list.

Consolidate the backup-file behavior into one place (Boundaries) and reference it from Purpose to remove the second mention.

Move the detailed Remove/Preserve/Compress rule tables into a references/ file (e.g., RULES.md) and keep a short summary in SKILL.md to improve progressive disclosure toward a 5.

Dimension	Reasoning	Score
Conciseness	The body is mostly lean and assumes Claude's competence — rule lists and pattern examples earn their place — with minor redundancy (the CRITICAL RULE re-states the code-block preservation already covered in 'Preserve EXACTLY', and the backup is mentioned in both Purpose and Boundaries).	4 / 5
Actionability	Gives a concrete runnable command ('cd caveman-compress && python3 -m scripts <absolute_filepath>') plus specific, actionable removal/preservation rule lists and before/after patterns; minor gaps (e.g., no inline flag reference) keep it just below copy-paste-complete 5.	4 / 5
Workflow Clarity	The numbered Process sequences the operation and explicitly embeds validation, a cherry-pick fix feedback loop, retry up to 2 times, and a leave-original-untouched fallback for the destructive overwrite — matching the anchor for clear sequence with explicit validation and error-recovery loops.	5 / 5
Progressive Disclosure	Well-organized into Purpose/Trigger/Process/Rules/Pattern/Boundaries with a clearly signaled, one-level reference to the scripts/ bundle (which exists), but the skill exceeds 50 lines and does not split detailed material into separate reference files, so it sits at 'good structure, minor gaps' rather than the 5 anchor.	4 / 5
	Total	17 / 20 Passed

Description

87%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.

A strong, third-person description that concretely states what the skill does and when to use it, with explicit trigger guidance and a distinct niche. Minor specificity and trigger-synonym gaps keep it just short of perfect on those sub-dimensions.

Dimension	Reasoning	Score
Specificity	Names the domain (natural language memory files) and lists several concrete actions — compress into caveman format, preserve technical substance/code/URLs/structure, overwrite original, save a backup as FILE.original.md — with only minor coverage gaps, fitting the 'several specific actions' anchor rather than the comprehensive 5.	4 / 5
Completeness	Explicitly answers 'what' (compress memory files into caveman format to save tokens while preserving substance) and 'when' (an explicit 'Trigger:' clause with concrete phrases), matching the anchor for clearly answering both with concrete triggers.	5 / 5
Trigger Term Quality	Provides both a slash command (/caveman-compress <filepath>) and a natural phrase ('compress memory file'), plus concrete file types (CLAUDE.md, todos, preferences), giving good keyword coverage though a few natural synonyms are missing.	4 / 5
Distinctiveness Conflict Risk	The 'caveman format' memory-file compression niche with its dedicated slash command and specific file targets is clearly distinct from other skills and unlikely to trigger for the wrong skill.	5 / 5
	Total	18 / 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: anthonystepvoy/caveman-opencode
Path: .opencode/skills/caveman-compress/SKILL.md
Commit: d3f5b6e

Reviewed: 22 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.