CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

caveman

Terse caveman voice: answer first, fluff gone, every technical fact kept. Use for /caveman, "caveman mode", "talk like caveman", "be brief", "less tokens". Stays on until "stop caveman" or "normal mode".

90

1.06x
Quality

90%

Does it follow best practices?

Impact

94%

1.06x

Average score across 3 eval scenarios

SecuritybySnyk

Passed

No findings from the security scan

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

SKILL.md
Quality
Evals
Security

Quality

Content

92%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 body is a model of what it preaches: dense, numbered rules with concrete bad/good pairs, explicit precedence and exception handling, and a pre-send validation checklist. The only nit is the short Why section, which mildly re-explains token economics Claude could infer.

DimensionReasoningScore

Conciseness

The body is exceptionally lean and rule-dense ("No greeting, hedging, pleasantries, recap, or closer"), but the Why section partially explains what Claude already knows ("'Sure, I'd be happy to help' is ten tokens. 'the' is one.") — minor instances that could be trimmed. Anchor 4 (efficient, minor over-explanation), not 5.

4 / 5

Actionability

Instruction-only skill with fully concrete guidance: Bad/Good example pairs ("Bad: 'Sure! I'd be happy to help…' / Good: 'Bug in auth middleware. Token expiry check use `<` not `<=`'"), exact banned-word lists, exact preservation rules ("Never drop not/never/no/only/except. Numbers and units exact."), and a copy-paste-analog pre-send checklist. Matches anchor 5; not 4 because the examples cover the common cases with no real gaps.

5 / 5

Workflow Clarity

Simple single-purpose skill where the action is unambiguous: rules are numbered, precedence is explicit ("Compression and clarity conflict? Clarity wins"), break-the-rules conditions are enumerated, and the 'Pre-send check' acts as a validation checklist before every response. Per the simple-skill guideline this qualifies for 5; no destructive or batch operations exist, so no cap applies.

5 / 5

Progressive Disclosure

Self-contained, well-organized sections (Persistence, Why, Rules, When to break the rules, Pre-send check) with nothing that belongs in a separate file; the only bundle files are two display SVG assets that correctly stay out of the body. Easy to navigate; anchor 5.

5 / 5

Total

19

/

20

Passed

Description

83%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: concrete capabilities, an explicit 'Use for…' trigger list, and even session lifecycle/exit conditions. The only weaknesses are a couple of missing natural synonyms and generic brevity triggers that slightly overlap with a general conciseness skill.

DimensionReasoningScore

Specificity

Names several concrete behaviors — "answer first, fluff gone, every technical fact kept" plus lifecycle behavior "Stays on until 'stop caveman' or 'normal mode'" — but coverage has minor gaps (e.g., no mention of exception/escape-hatch behavior or that it applies to every response). Fits anchor 4 (several specific actions, minor gaps), not 5 (not comprehensive) and not 3 (well beyond 1-2 actions).

4 / 5

Completeness

Explicitly answers both: what ("Terse caveman voice: answer first, fluff gone, every technical fact kept") and when ("Use for /caveman, 'caveman mode', 'talk like caveman', 'be brief', 'less tokens'") with concrete trigger phrases, plus exit conditions. Matches anchor 5 exactly; not 4 since the 'when' clause is fully explicit with specific triggers.

5 / 5

Trigger Term Quality

Includes the slash command (/caveman) and natural user phrases "caveman mode", "talk like caveman", "be brief", "less tokens" — good keyword coverage that clearly beats the single-domain anchor 3. Not 5 because common synonyms like "concise", "shorter answers", or "stop being verbose" are absent.

4 / 5

Distinctiveness Conflict Risk

The "caveman" niche with triggers like /caveman and "talk like caveman" is highly distinctive, but the generic brevity triggers "be brief" and "less tokens" create minor overlap risk with any general conciseness-style skill — anchor 4 (mostly distinct, minor overlap), not 5.

4 / 5

Total

17

/

20

Passed

Validation

100%

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/SKILL.md

Reviewed

Table of Contents

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.