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

81%

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

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

Quality

Content

75%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 skill body with a clear destructive-operation workflow that includes validation and retry feedback loops, and appropriately externalizes implementation to a real scripts/ bundle. Slight verbosity in examples and delegation of validation criteria keep most dimensions at 4 rather than 5.

Suggestions

Trim the redundant second original/compressed example pair in the Pattern section to reduce tokens, since both illustrate the same transformation.

State the validation criteria the CLI applies (what 'validate output' checks) inline or via a one-line pointer to scripts/validate.py, so the workflow's validation step is self-contained.

Add the explicit retry/fix command form (e.g., how a targeted fix is invoked) so the recovery branch is copy-paste executable, not only described.

Dimension	Reasoning	Score
Conciseness	Lean bullet structure with no padding or explanation of concepts Claude already knows; the second full original/compressed example pair is somewhat redundant and could be trimmed, keeping it just below a 5.	4 / 5
Actionability	Provides a copy-paste command ("python3 -m scripts <absolute_filepath>") and concrete, exhaustive remove/preserve/boundary rules; minor gaps such as the retry invocation details being delegated to the CLI keep it from a 5.	4 / 5
Workflow Clarity	Numbered process with explicit validation ("validate output") and a feedback loop ("cherry-pick fix", "retry up to 2 times", "if still failing ... leave original file untouched") for a destructive overwrite operation; the validation criteria themselves are delegated to scripts rather than stated, a minor gap.	4 / 5
Progressive Disclosure	Well-organized sections (Purpose, Trigger, Process, Compression Rules, Pattern, Boundaries) with a real one-level-deep bundle reference ("scripts live in scripts/", "search for scripts/__main__.py") that maps to actual files; bulk logic is appropriately externalized to scripts with only minor navigation gaps.	4 / 5
	Total	16 / 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 specific, third-person description that clearly states what the skill does and when to invoke it via an explicit trigger clause. Trigger-term and action coverage are strong, with only minor gaps in named actions.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions ("Compress natural language memory files into caveman format", "Preserves all technical substance, code, URLs, and structure", "overwrites the original file", "backup saved as FILE.original.md"), with only minor gaps such as omitting detect/validate from the description.	4 / 5
Completeness	Explicitly answers both what (compress memory files into caveman format, preserve substance, overwrite, backup) and when ("Trigger: /caveman-compress FILEPATH or 'compress memory file'"), with concrete trigger phrases, so it is not capped.	5 / 5
Trigger Term Quality	Good natural-term coverage ("compress memory file", "CLAUDE.md, todos, preferences", "/caveman-compress FILEPATH") a user would plausibly say, though a few synonyms or alternate phrasings are missing.	4 / 5
Distinctiveness Conflict Risk	"caveman format" memory-file compression is a clear niche with a distinctive named trigger and command form, giving minimal conflict risk with other skills.	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: masweb/waxp-vuebo
Path: .agents/skills/caveman-compress/SKILL.md
Commit: 286c58f

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.