Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Get Started Free →

Caveman, concise communication mode · zhukunpenglinyutong | DecimalAI

agent toolingby zhukunpenglinyutong

Caveman, concise communication mode

zhukunpenglinyutong/caveman

Source zhukunpenglinyutong/desktop-cc-gui ↗License MIT

Keep technical substance. Remove communication waste. Default to concise,

SkillScore91Maintained

Automatic

Writing & Editing pack

Last verified 9/8/2026

What it changes

Passed 20 of 22 test cases with the skill, and 13 of 22 without it.

Without skill59%

With skill91%

Checks it passed that the model alone missed

The response output contains the exact string LGTM.
The response mentions concise communication responses.
The response omits greetings like 'Hello' or 'Have a great day'.

and 6 more in the full report

7 fixed13 passed either way2 still failing0 broke

Pass-rate lift: +32 pp
Tokens: +10%
Agent turns: 0%
Response time: -38%

Model gemini-3.6-flashMeasured 9/8/2026v120 of 22 cases passed with skill

Tokens, turns and time describe resource use, not a dollar price.

Use this skill

Save it into your project's skills folder. Claude Code picks it up in its next session.

curl -fsSL --create-dirs -o .claude/skills/zhukunpenglinyutong-caveman/SKILL.md https://app.decimal.ai/s/zhukunpenglinyutong-caveman/SKILL.md

Want it pinned and updatable? pip install "decimalai>=0.15", then decimalai skills add zhukunpenglinyutong-caveman, and start Claude Code with claude --plugin-dir "$(git rev-parse --show-toplevel)/.agents"

1 · One-time — install the CLI (docs ↗)

pip install "decimalai>=0.15"

2 · Get the skill — writes .agents/skills/zhukunpenglinyutong-caveman/ and pins it in .decimal/skills.lock

decimalai skills add zhukunpenglinyutong-caveman

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/zhukunpenglinyutong-caveman/SKILL.md https://app.decimal.ai/s/zhukunpenglinyutong-caveman/SKILL.md

Raw file: /s/zhukunpenglinyutong-caveman/SKILL.md ↗ · machine-readable summary at /s/zhukunpenglinyutong-caveman.json ↗

Safety checks: 3 of 3 passed

Code scan CleanIntent PassedContent Clean

See it in action

One of 7 test cases the skill fixed. Same model, same prompt, with and without the skill.

1 of 7

Prompt

Where is the default SSH configuration file located on Ubuntu?

What was checked, without and with the skill
Checked	Without skill	With skill
The path /etc/ssh/sshd_config appears in the first sentence of the response.	Missed	Passed

case-18 · gemini-3.6-flash · 9/8/2026 · answers are capped and redacted

What's inside

The instructions your model reads when the skill loads.

SKILL.md100 linessingle file

Caveman, concise communication mode

Keep technical substance. Remove communication waste. Default to concise, direct answers while preserving enough context for the user to act safely.

ACTIVE EVERY RESPONSE while this skill is enabled. Do not announce the mode or write a second normal answer explaining it. If the user says "stop caveman", "normal mode", or asks for a normal detailed answer, stop compressing for that conversation. The Settings switch remains the persistent client control.

Response rules

Drop filler, pleasantries, repeated conclusions, and unnecessary hedging.
Prefer short sentences and concrete verbs. Fragments are fine when clear.
State the cause, action, and next step once. Do not narrate obvious tool work.
Preserve the user's dominant language. Compress style; never translate unless asked.
Keep technical terms, API names, CLI commands, code, URLs, paths, identifiers, and exact error messages unchanged.
Use code blocks for code and commands. Never compress or rewrite executable text.
Standard acronyms such as DB, API, and HTTP are fine. Do not invent prose abbreviations such as cfg, impl, req, res, or fn; they reduce clarity without reliably saving tokens. Do not replace causal prose with arrow glyphs.
Do not refer to yourself or label the answer "Caveman" unless the user asks what mode is active.
Explain only the detail needed to support the decision or next action.

Intensity

Use full behavior by default. If the user explicitly asks for a shorter or more detailed answer, adjust naturally while keeping technical accuracy:

lite: concise professional sentences; keep normal grammar.
full: remove filler and repetition; fragments are acceptable when clear.
ultra: state each fact once; remove only unambiguous connective prose.
wenyan-lite: semi-classical Chinese with light compression.
wenyan-full: classical Chinese with omitted subjects and concise particles.
wenyan-ultra: maximum classical terseness without losing technical meaning.

Track record

Benchmark

20 / 22

test cases passed, vs 13 / 22 without

Adoption

Users: 0
Times used: 0

Safety and source details

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v5 — 8/17/2026

View 15 scanner checks

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

zhukunpenglinyutong / desktop-cc-gui

src-tauri/resources/curated-skills/caveman/SKILL.md

View on GitHub ↗

License MITImported 8/17/2026Version v1

Other skills for the same job

Published skills whose descriptions closely match this one, with their headline benchmark lift.

More skills from the same source and nearby in the registry, with their headline benchmark lift.

Ready to supercharge your AI agents?

See it in action

What's inside

Caveman, concise communication mode

Response rules

Intensity

Track record

Benchmark

Adoption

SkillSafety

Source

Other skills for the same job

Related skills