Local Agent Workspace
Description
🧠 C.A.S.E. Hybrid Cloud+Local AI Framework ── 雲端 AI 出計畫,本地 AI 執行苦勞。Llama.cpp 部署實戰指南,資料物理隔離,讓 Token 花在刀口上、不受 Quota 所困。
Installation
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open the source below and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
README
🚀 Local-Agent-Workspace
[!IMPORTANT] **個人立場聲明:** 本專案僅為個人技術研究分享,所有內容與參數調校均基於公開開源數據。專案內容不代表任何機關立場,亦不涉及任何公務機敏資料。
本地 AI 極致壓榨與開發規範雙軌指南 (Hardware & Software Dual-Track Guide)
本專案提供兩大獨立且可平行參考的本地 AI 實戰維度:
graph TD
A["本地 AI 開發實戰專案 (Local-Agent-Workspace)"] --> B["🟢 軌道一:硬體選型與算力壓榨 (Hardware Track)"]
A --> C["🔵 軌道二:軟體框架與 AI 規範 (Software Track)"]
B --> B1["選擇硬體平台 -> 複製配置啟動本地 Server"]
C --> C1["一鍵植入 C.A.S.E 規則手冊 -> AI Agent 自動建立專案規範"]
🟢 軌道一:硬體選型與本機算力極致壓榨 (Hardware Track)
📊 本地算力平台快速選取看板
| 硬體環境 (Hardware Platform) | 核心推薦模型 (Recommended Model) | 檔案大小 (Size) | 推理效能 (Inference Performance) |
|---|---|---|---|
| 高階顯卡 (20GB+ VRAM) | GRM-2.6-Opus 27B / Qwopus 27B | 15.3G / 15.4G | MTP 投機解碼 (~49 T/s) |
| 中階顯卡 (16GB VRAM) | Qwen3.6-35B-A3B-Cerebellum | 12 GB | GPU 全卸載 MoE 線性推理 |
| 純 CPU / 大 RAM (32GB+) | Qwen3.6-35B-A3B-Cerebellum | 12 GB | MoE+SSM 純 CPU 線性推理 |
📦 運算引擎與模型權重下載 (Llama.cpp & Models Download)
- ⚡ 推薦一鍵自動更新:雙擊本專案根目錄的
run-update.bat即可全自動下載最新版llama.cpp並完成解壓合併。- 📥 一鍵下載更新腳本 (免克隆專案):於目標安裝資料夾開啟終端機,貼上執行以下指令即可直接取得更新雙檔:
- PowerShell:
Invoke-WebRequest -Uri "https://raw.githubusercontent.com/Chiakai-Chang/Local-Agent-Workspace/main/run-update.bat" -OutFile "run-update.bat"; Invoke-WebRequest -Uri "https://raw.githubusercontent.com/Chiakai-Chang/Local-Agent-Workspace/main/update-llama-cpp.ps1" -OutFile "update-llama-cpp.ps1" - cURL (Git Bash/Linux):
curl -fsSL https://raw.githubusercontent.com/Chiakai-Chang/Local-Agent-Workspace/main/run-update.bat -o run-update.bat && curl -fsSL https://raw.githubusercontent.com/Chiakai-Chang/Local-Agent-Workspace/main/update-llama-cpp.ps1 -o update-llama-cpp.ps1
- PowerShell:
- ⚠️ 首次使用:請先以文字編輯器開啟
update-llama-cpp.ps1,將第 6 行的$TargetDir = "D:\MyProject\llama"修改為您本機實際要安裝的資料夾路徑!
- 📥 一鍵下載更新腳本 (免克隆專案):於目標安裝資料夾開啟終端機,貼上執行以下指令即可直接取得更新雙檔:
- 📦 備用手動下載:前往 [Llama.cpp Releases](https://github.com/ggml-org/llama.cp
Related Skills
Agency Agents
A complete AI agency at your fingertips - From frontend wizards to Reddit community ninjas, from whimsy inject
AI Firecrawl
🔥 The API to search, scrape, and interact with the web for AI
AI Artifacts Builder
Suite of tools for creating elaborate, multi-component claude.ai HTML artifacts using modern frontend web tech
AI CrewAI
Framework for orchestrating role-playing, autonomous AI agents. By fostering collaborative intelligence, CrewA
AI TrendRadar
⭐AI-driven public opinion & trend monitor with multi-platform aggregation, RSS, and smart alerts.🎯 告别信息过载,你的
AI mem0
| Universal memory layer for AI Agents | 51341 | 221 | 1 |
AI