Name: caveman
Rating: 76 (1 reviews)
Author: alirezarezvani

caveman

Ultra-compressed communication mode. Cuts token usage ~75% by dropping filler, articles, and pleasantries while keeping full technical accuracy. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman.

Quality

95%

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 in alirezarezvani/claude-skills

Quality

Content

96%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.

An exemplary terse skill body: dense, actionable, and well-structured. The only real defect is a broken reference — the body points to references/companion_tooling.md, but no references/ directory or file ships with the skill.

Suggestions

Ship the referenced references/companion_tooling.md (or remove the link) so the one external reference resolves; a broken link is the sole thing holding progressive_disclosure below 5.

Clarify in the Tooling section whether cs-caveman-mode / /cs:caveman are bundled commands or external prerequisites, so the guidance is self-contained even if the reference is missing.

Add a one-line 'How to stop' trigger alongside the existing 'stop caveman'/'normal mode' phrasing in the body's Persistence section to mirror the description's explicit off-switch.

Dimension	Reasoning	Score
Conciseness	Lean and efficient to the point of dogfooding its own brevity mandate — no explaining of concepts Claude already knows; every line (rules, pattern, before/after examples) earns its place.	5 / 5
Actionability	Concrete, copy-paste-style guidance: explicit Drop/Keep lists (articles, filler, abbreviations like DB/auth/config), a precise output pattern '[thing] [action] [reason]. [next step].', and worked before/after examples covering common cases.	5 / 5
Workflow Clarity	A simple single-purpose style skill with an unambiguous action; sections sequence trigger→rules→examples→exception→resume, and the Auto-Clarity Exception even models a validation checkpoint ('Verify backup exist first') for the illustrative destructive case, satisfying the simple-skill exception.	5 / 5
Progressive Disclosure	Well-organized sections under 50 lines with a one-level-deep, clearly signaled pointer to references/companion_tooling.md; held at 4 rather than 5 because that referenced file does not exist in the bundle, making the navigation link broken.	4 / 5
	Total	19 / 20 Passed

Description

95%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 description: concise, third-person, with explicit what/when structure and a rich set of natural trigger phrases. Minor specificity gap only because the actions are stylistic compressions rather than a broad capability set.

Dimension	Reasoning	Score
Specificity	Names the domain ('Ultra-compressed communication mode') and several concrete actions ('Cuts token usage ~75% by dropping filler, articles, and pleasantries while keeping full technical accuracy'), matching the 'lists several specific actions; minor gaps' anchor; not a 5 because the actions are compression heuristics rather than a comprehensive multi-capability list.	4 / 5
Completeness	Explicitly answers both what ('Ultra-compressed communication mode... Cuts token usage ~75%') and when ('Use when user says "caveman mode"...') with concrete trigger phrases.	5 / 5
Trigger Term Quality	Comprehensive natural triggers a user would actually say — 'caveman mode', 'talk like caveman', 'use caveman', 'less tokens', 'be brief' — plus the slash command '/caveman', covering synonyms and the invocation form.	5 / 5
Distinctiveness Conflict Risk	'Caveman mode' is a clear niche with distinct, idiosyncratic triggers and minimal overlap risk with other skills.	5 / 5
	Total	19 / 20 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
relative_links	Relative link issues: 2 missing	Warning
referenced_paths_exist	Referenced path issues: 4 missing	Warning

	Total	14 / 16 Passed

Repository: alirezarezvani/claude-skills
Path: .gemini/skills/caveman/SKILL.md
Commit: 19392f7

Reviewed: ~2 hours 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.