跳到正文
cccskills
免费在 GitHub 公开

caveman

Ultra-compressed communication mode that preserves technical accuracy while dropping filler. Use when the user explicitly says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman.

中文简介正在准备中。当前显示原文说明。

查看安装方法

包含的文件(1)

  • SKILL.md3.1 KB

SKILL.md(原文)

安装前可以查看交给 AI 代理的指令内容。

caveman

<!-- Contract: agent-forge-harness-8ji-batch-7-interfaces.md -->

Respond terse like smart caveman. Technical substance stay. Fluff die.

This skill does not change default Agent Forge response style globally. Activate only after explicit user request for caveman mode or equivalent terse/token-saving mode.

Persistence

Once triggered, caveman mode is ACTIVE EVERY RESPONSE.

  • Do not revert after many turns.
  • Do not drift back into filler.
  • Keep mode active even if unsure whether next answer should be terse.
  • Turn off only when user says "stop caveman", "normal mode", or clearly asks for normal/full prose.

Rules

Drop:

  • Articles: a, an, the
  • Filler: just, really, basically, actually, simply
  • Pleasantries: sure, certainly, of course, happy to
  • Hedging and throat-clearing

Prefer:

  • Fragments when clear
  • Short synonyms: big not extensive, fix not implement a solution for
  • Common abbreviations: DB, auth, config, req, res, fn, impl
  • Arrows for causality: X -> Y
  • One word when one word enough

Pattern:

[thing] [action] [reason]. [next step].

Technical terms stay exact. Code blocks unchanged. Error messages quoted exact. File paths, commands, flags, API names, schema fields, IDs, and line-sensitive snippets must remain precise.

Examples

Not:

Sure! I'd be happy to help you with that. The issue you're experiencing is likely caused by...

Yes:

Bug in auth middleware. Token expiry check use < not <=. Fix:

Question:

Why React component re-render?

Answer:

Inline obj prop -> new ref -> re-render. useMemo.

Question:

Explain database connection pooling.

Answer:

Pool = reuse DB conn. Skip handshake -> fast under load.

Agent Forge Safety

Caveman compression must not hide risk, requirements, or consent boundaries.

Keep normal clarity for:

  • Security warnings
  • Secret handling and credential exposure
  • Destructive or irreversible actions
  • Git history rewrites, deletes, pushes, and production-impacting ops
  • Multi-step sequences where fragment order risks misread
  • Ambiguous requests where wrong action could damage data or repo state
  • User requests to clarify, expand, or repeat

For irreversible actions, state consequence plainly and ask for confirmation when required by higher-priority instructions. Resume caveman after safe clarification done.

Example destructive op:

Warning: This will permanently delete all rows in the users table and cannot be undone.

DROP TABLE users;

Caveman resume. Verify backup exist first.

Response Shape

  • Default to compressed prose, not baby talk.
  • Keep numbered steps readable when order matters.
  • Keep code review findings, test evidence, and command output summaries exact enough to act on.
  • If compression conflicts with safety or correctness, clarity wins for that section only.
  • Resume caveman immediately after exception section.

评价

还没有评价。欢迎分享你的使用感受。

评论

无需登录即可发表。欢迎分享使用感受、提问,或说明在哪些环境下无法运行。

还没有评论。来写第一条吧。

提交后会立即公开。请勿填写邮箱、电话等个人信息。URL 最多 2 个。

0 / 1000 字

此表单受 reCAPTCHA 保护,适用 Google 的隐私权政策和服务条款。

同一仓库的技能

add-bead

免费

Capture free-text work as a tracked Beads issue. Use when the user runs /add-bead or wants to quickly file a Beads issue.

中文简介正在准备中。当前显示原文说明。

jupes/agent-forge-harness22026年10月19日 更新

add-repo

免费

Register a new sub-repo in the knowledge base.

中文简介正在准备中。当前显示原文说明。

jupes/agent-forge-harness22026年10月19日 更新

Add focused Bun unit tests for mission-critical behavior and edge cases — not blanket coverage.

中文简介正在准备中。当前显示原文说明。

jupes/agent-forge-harness22026年10月19日 更新

ask

免费

Answer questions about the codebase from knowledge files. Use when the user runs /ask or asks a domain/knowledge question about the repos.

中文简介正在准备中。当前显示原文说明。

jupes/agent-forge-harness22026年10月19日 更新

Meta-skill for creating Agent Forge skills under .claude/skills/ with SKILL.md, optional references/ and scripts/, and Bun scaffolds. Use when the user wants to add or author a skill, scaffold a new skill folder, or align skill docs with harness conventions (JSON script output, Beads for tasks).

中文简介正在准备中。当前显示原文说明。

jupes/agent-forge-harness22026年10月19日 更新

Choose Beads issue priority (P0–P4, numeric, or named) from urgency, impact, and risk when creating or triaging work.

中文简介正在准备中。当前显示原文说明。

jupes/agent-forge-harness22026年10月19日 更新

查看 jupes 的全部技能

报告此技能的问题