冷啟新 session:掃描 codebase 結構並讀取 README,快速建立專案 context。適合在陌生 repo 或新 session 開始時使用。
Scanned 6/5/2026
Install to Claude Code
npx -y skills add zeuikli/claude-code-workspace --skill prime --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Prime?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/zeuikli-prime)More formats (shields.io, HTML) on the badges page.
---
name: prime
description: 冷啟新 session:掃描 codebase 結構並讀取 README,快速建立專案 context。適合在陌生 repo 或新 session 開始時使用。
disable-model-invocation: true
context: fork
---
# Prime — 冷啟 Context Loading
## 觸發條件
- 新 session 開始,需要快速了解一個不熟悉的 codebase
- 接手他人專案或剛 clone 的 repo
- 從 `/clear` 或 `/compact` 後重新建立 context
## 執行步驟
1. **掃描 git 追蹤的檔案結構**
```bash
git ls-files | head -80
```
2. **取得目錄樹概覽**
```bash
find . -type f -not -path './.git/*' -not -path './node_modules/*' | head -60 | sort
```
3. **讀取專案文件**(按優先順序)
- `@README.md`(若存在)
- `@CLAUDE.md`(若存在)
- `@docs/INDEX.md`(若存在)
4. **產出 Context 摘要**,包含:
- 專案目的(1-2 句)
- 技術棧
- 主要模組/目錄說明
- 下一步建議(適合開始什麼工作)
## 針對本 workspace 的額外步驟
若在 `claude-code-workspace` 中執行,額外讀取:
- `.claude/rules/` 目錄列表(了解現有規則)
- `.claude/skills/` 目錄列表(了解現有 skills)
- `.claude/agents/` 目錄列表(了解現有 agents)
- `scripts/healthcheck.sh` 提示可執行健康檢查
## Gotcha
- **不要** 讀取超過 3 個文件全文,優先讀入 README + CLAUDE.md 即可
- 若 README 超過 200 行,只讀前 80 行摘要
- context 摘要長度控制在 ≤300 字,避免佔用過多 context budget
No comments yet. Be the first to comment!
Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...