admin/Agentic-Compaction-and-Pipleline-by-GLM-5
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat: Add complete Agentic Compaction & Pipeline System

Browse Source 
- Context Compaction System with token counting and summarization
- Deterministic State Machine for flow control (no LLM decisions)
- Parallel Execution Engine (up to 12 concurrent sessions)
- Event-Driven Coordination via Event Bus
- Agent Workspace Isolation (tools, memory, identity, files)
- YAML Workflow Integration (OpenClaw/Lobster compatible)
- Claude Code integration layer
- Complete demo UI with real-time visualization
- Comprehensive documentation and README

Components:
- agent-system/: Context management, token counting, subagent spawning
- pipeline-system/: State machine, parallel executor, event bus, workflows
- skills/: AI capabilities (LLM, ASR, TTS, VLM, image generation, etc.)
- src/app/: Next.js demo application

Total: ~100KB of production-ready TypeScript code
This commit is contained in:
Z User
2026-03-03 12:40:47 +00:00
Unverified
parent 63a8b123c9
commit 2380d33861
152 changed files with 51569 additions and 817 deletions
Download Patch File Download Diff File Expand all files Collapse all files

83
skills/gift-evaluator/SKILL.md Executable file
View File

@@ -0,0 +1,83 @@
---
name: gift-evaluator
description: The PRIMARY tool for Spring Festival gift analysis and social interaction generation. Use this skill when users upload photos of gifts (alcohol, tea, supplements, etc.) to inquire about their value, authenticity, or how to respond socially. Integrates visual perception, market valuation, and HTML card generation.
license: Internal Tool
---
This skill transforms the assistant into an "AI Gift Appraiser" (春节礼品鉴定师). It bridges the gap between raw visual data and complex social context. It is designed to handle the full lifecycle of a user's request: identifying the object, determining its market and social value, and producing a shareable, gamified HTML artifact.
## Agent Thinking Strategy
Before and during the execution of tools, maintain a "High EQ" and "Market-Savvy" mindset. You are not just identifying objects; you are decoding social relationships.
1. **Visual Extraction (The Eye)**:
* Call the vision tool to get a raw description.
* **CRITICAL**: Read the raw description carefully. Extract specific entities: Brand names (e.g., "Moutai", "Dior"), Vintages, Packaging details (e.g., "Dusty bottle" implies old stock, "Gift box" implies formality).
2. **Valuation Logic (The Brain)**:
* **Price Anchoring**: Use search tools to find the *current* market price.
* **Social Labeling**: Classify the gift based on price and intent:
* `luxury`: High value (> ¥1000), "Hard Currency".
* `standard`: Festive, safe choices (¥200 - ¥1000).
* `budget`: Practical, funny, or cheap (< ¥200).
3. **Creative Synthesis (The Mouth)**:
* **Deep Critique**: Generate a "Roast" (毒舌点评) of **at least 50 words**. It must combine the visual details (e.g., dust, packaging color) with the price reality. Be spicy but insightful.
* **Structured Strategy**: You must structure the "Thank You Notes" and "Return Gift Ideas" into JSON format for the UI to render.
## Tool Usage Guidelines
### 1. The Perception Phase (Visual Analysis)
Purpose: Utilizing VLM skills to conduct a multi-dimensional visual decomposition of the uploaded product image. This process automatically identifies and extracts structured data including Brand Recognition, Product Style, Packaging Design, and Aesthetic Category.
**Output Analysis**:
* The tool returns a raw string content. Read it to extract keywords for the next step.
### 2. The Valuation Phase (Search)
**Purpose**: Validate the product's worth.
**Command**:search "EXTRACTED_KEYWORDS + price + review"
### 3. The Content Structuring Phase (Reasoning)
**Purpose**: Prepare the data for the HTML generator. **Do not call a tool here, just think and format strings.**
1. **Construct `thank_you_json**`: Create 3 distinct styles of private messages.
* *Format*: `[{"style": "Style Name", "content": "Message..."}]`
* *Requirement*:
* Style 1: "Decent/Formal" (for elders/bosses).
* Style 2: "Friendly/Warm" (for peers/relatives).
* Style 3: "Humorous/Close" (for best friends).
2. **Construct `return_gift_json**`: Analyze 4 potential giver personas.
* *Format*: `[{"target": "If giver is...", "item": "Suggest...", "reason": "Why..."}]`
* *Requirement*: Suggestions must include Age/Gender/Relation analysis (e.g., "If giver is an elder male", "If giver is a peer female").
* *Value Logic*: Adhere to the principle of Value Reciprocity. The return gift's value should primarily match the received gift's value, while adjusting slightly based on the giver's status (e.g., seniority or intimacy).
### 4. The Creation Phase (Render)
**Purpose**: Package the analysis into a modern, interactive HTML card.
**HTML Generation**:
* *Constraint*: The `image_url` parameter in the Python command MUST be the original absolute path.`output_path` must be the full path.
* *Command*:
```bash
python3 html_tools.py generate_gift_card \
--product_name "EXTRACTED_NAME" \
--price "ESTIMATED_PRICE" \
--evaluation "YOUR_LONG_AND_SPICY_CRITIQUE" \
--thank_you_json '[{"style":"...","content":"..."}]' \
--return_gift_json '[{"target":"...","item":"...","reason":"..."}]' \
--vibe_code "luxury|standard|budget" \
--image_url "IMAGE_FILE_PATH" \
--output_path "TARGET_FILE_PATH"
```
## Operational Rules
1. **JSON Formatting**: The `thank_you_json` and `return_gift_json` arguments MUST be valid JSON strings using double quotes. Do not wrap them in code blocks inside the command.
2. **Critique Depth**: The `evaluation` text must be rich. Don't just say "It's expensive." Say "This 2018 vintage shows your uncle raided his personal cellar; the label wear proves it's real."
3. **Vibe Consistency**: Ensure `vibe_code` matches the `price` assessment.
4. **Final Output**: Always present the path to the generated HTML file.

268
skills/gift-evaluator/html_tools.py Executable file
View File

@@ -0,0 +1,268 @@
import os
import argparse
import json
import html
import base64
import mimetypes
import urllib.request
def generate_gift_card(product_name, price, evaluation, thank_you_json, return_gift_json, vibe_code, image_url, output_path="gift_card_result.html"):
"""
生成现代风格的交互式礼品鉴定卡片。
"""
# --- 图片转 Base64 逻辑 (保持上一步功能) ---
final_image_src = image_url
try:
image_data = None
mime_type = None
if image_url.startswith(('http://', 'https://')):
req = urllib.request.Request(image_url, headers={'User-Agent': 'Mozilla/5.0'})
with urllib.request.urlopen(req, timeout=10) as response:
image_data = response.read()
mime_type = response.headers.get_content_type()
else:
if os.path.exists(image_url):
mime_type, _ = mimetypes.guess_type(image_url)
with open(image_url, "rb") as f:
image_data = f.read()
if image_data:
if not mime_type: mime_type = "image/jpeg"
b64_str = base64.b64encode(image_data).decode('utf-8')
final_image_src = f"data:{mime_type};base64,{b64_str}"
except Exception as e:
print(f"⚠️ 图片转换 Base64 失败，使用原链接。错误: {e}")
# --- 1. 数据解析 ---
try:
thank_you_data = json.loads(thank_you_json)
except:
thank_you_data = [{"style": "通用版", "content": thank_you_json}]
try:
return_gift_data = json.loads(return_gift_json)
except:
return_gift_data = [{"target": "通用建议", "item": return_gift_json, "reason": "万能回礼"}]
# --- 2. 风格配置 ---
styles = {
"luxury": {
"page_bg": "bg-neutral-900",
"card_bg": "bg-neutral-900/80 backdrop-blur-xl border border-white/10",
"text_main": "text-white", "text_sub": "text-neutral-400",
"accent": "text-amber-400", "tag_bg": "bg-amber-400/20 text-amber-400",
"btn_hover": "hover:bg-amber-400 hover:text-black",
"img_bg": "bg-neutral-800" # 图片衬底色
},
"standard": {
"page_bg": "bg-stone-200",
"card_bg": "bg-white/95 backdrop-blur-xl border border-stone-200",
"text_main": "text-stone-800", "text_sub": "text-stone-500",
"accent": "text-red-600", "tag_bg": "bg-red-50 text-red-600",
"btn_hover": "hover:bg-red-600 hover:text-white",
"img_bg": "bg-stone-100"
},
"budget": {
"page_bg": "bg-yellow-50",
"card_bg": "bg-white border-4 border-black shadow-[8px_8px_0px_0px_rgba(0,0,0,1)]",
"text_main": "text-black", "text_sub": "text-gray-600",
"accent": "text-blue-600", "tag_bg": "bg-black text-white",
"btn_hover": "hover:bg-blue-600 hover:text-white",
"img_bg": "bg-gray-200"
}
}
st = styles.get(vibe_code, styles["standard"])
if "img_bg" not in st: st["img_bg"] = "bg-black/5" # 兼容兜底
# --- 3. 辅助逻辑 ---
is_dark_mode = "text-white" in st['text_main']
bubble_bg = "bg-white/10 border-white/10" if is_dark_mode else "bg-black/5 border-black/5"
bubble_hover = "hover:bg-white/20" if is_dark_mode else "hover:bg-black/10"
divider_color = "border-white/20" if is_dark_mode else "border-black/10"
# --- 4. HTML 构建 ---
thank_you_html = ""
for item in thank_you_data:
thank_you_html += f"""
<div class="group relative p-4 rounded-xl {bubble_bg} border {bubble_hover} transition-all cursor-pointer mb-3" onclick="copyText(this, '{html.escape(item['content'], quote=True)}')">
<div class="flex justify-between items-center mb-2">
<span class="text-xs font-bold {st['accent']} border border-current px-2 py-0.5 rounded-full">{item['style']}</span>
<span class="text-[10px] opacity-60 group-hover:opacity-100 transition-opacity {st['text_sub']} flex items-center gap-1">
<svg class="w-3 h-3" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M8 16H6a2 2 0 01-2-2V6a2 2 0 012-2h8a2 2 0 012 2v2m-6 12h8a2 2 0 002-2v-8a2 2 0 00-2-2h-8a2 2 0 00-2 2v8a2 2 0 002 2z"></path></svg>
点击复制
</span>
</div>
<p class="text-sm {st['text_main']} leading-relaxed opacity-95 font-medium">{item['content']}</p>
<div class="copy-feedback absolute inset-0 bg-{st['accent'].split('-')[1]}-500 text-white flex items-center justify-center rounded-xl opacity-0 pointer-events-none transition-opacity duration-200 font-bold z-10">
<span>✓ 已复制</span>
</div>
</div>
"""
return_gift_html = ""
for item in return_gift_data:
return_gift_html += f"""
<div class="p-4 rounded-xl {bubble_bg} border flex flex-col justify-between h-full hover:scale-[1.02] transition-transform duration-300">
<div class="flex items-center gap-2 mb-2">
<div class="w-1.5 h-1.5 rounded-full bg-current {st['accent']}"></div>
<div class="text-xs font-bold uppercase tracking-wider {st['text_sub']}">{item['target']}</div>
</div>
<div class="font-bold {st['text_main']} text-lg mb-2">{item['item']}</div>
<div class="text-xs {st['text_sub']} opacity-80 leading-snug bg-black/5 dark:bg-white/5 p-2 rounded">{item['reason']}</div>
</div>
"""
html_content = f"""
<!DOCTYPE html>
<html lang="zh-CN">
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>礼品鉴定报告</title>
<script src="https://cdn.tailwindcss.com"></script>
<link href="https://fonts.googleapis.com/css2?family=Inter:wght@400;600;800&family=Noto+Serif+SC:wght@700;900&display=swap" rel="stylesheet">
<style>
body {{ font-family: 'Inter', sans-serif; }}
.serif {{ font-family: 'Noto Serif SC', serif; }}
/* 通用滚动条样式 */
.custom-scroll::-webkit-scrollbar {{ width: 4px; }}
.custom-scroll::-webkit-scrollbar-thumb {{ background-color: rgba(150,150,150,0.3); border-radius: 4px; }}
.custom-scroll:hover::-webkit-scrollbar-thumb {{ background-color: rgba(150,150,150,0.6); }}
</style>
</head>
<body class="{st['page_bg']} min-h-screen flex items-center justify-center p-2 md:p-8 selection:bg-red-200 selection:text-red-900">
<div class="w-full max-w-6xl {st['card_bg']} rounded-[2.5rem] shadow-2xl overflow-hidden relative flex flex-col md:flex-row md:h-[750px] transition-all duration-500">
<div class="w-full md:w-[45%] flex flex-col relative shrink-0 border-b md:border-b-0 md:border-r {divider_color}">
<div class="relative h-72 md:h-[55%] group overflow-hidden {st['img_bg']} flex items-center justify-center p-6">
<img src="{final_image_src}" class="w-full h-full object-contain relative z-10 drop-shadow-xl transition-transform duration-700 group-hover:scale-105">
<div class="absolute inset-x-0 bottom-0 h-32 bg-gradient-to-t from-black/80 to-transparent z-20 pointer-events-none"></div>
<div class="absolute bottom-6 left-6 right-6 z-30">
<div class="inline-block px-3 py-1 rounded-lg text-[10px] font-bold uppercase tracking-widest mb-2 {st['tag_bg']} backdrop-blur-md shadow-lg">
AI Gift Analysis
</div>
<h1 class="text-3xl md:text-4xl font-black text-white leading-tight serif mb-1 drop-shadow-md truncate">{product_name}</h1>
<div class="flex items-baseline gap-2 text-white/90">
<span class="text-sm font-light opacity-80">当前估值</span>
<span class="text-3xl font-bold tracking-tight">{price}</span>
</div>
</div>
</div>
<div class="flex-1 p-6 md:p-8 flex flex-col min-h-0 bg-inherit relative">
<div class="absolute top-0 left-8 -mt-5 text-6xl opacity-20 {st['text_main']} font-serif select-none">“</div>
<h3 class="text-xs font-bold uppercase tracking-widest {st['text_sub']} mb-3 flex items-center gap-2 shrink-0">
<span class="w-8 h-[1px] bg-current opacity-50"></span>
专家鉴定评价
</h3>
<div class="{st['text_main']} text-base md:text-lg leading-relaxed italic font-medium relative z-10 overflow-y-auto custom-scroll flex-1 pr-2">
{evaluation}
</div>
<div class="mt-4 pt-4 border-t {divider_color} flex items-center gap-3 shrink-0">
<div class="w-8 h-8 rounded-full {st['tag_bg']} flex items-center justify-center font-bold text-xs">AI</div>
<div class="flex flex-col">
<span class="text-xs font-bold {st['text_main']}">首席鉴定官</span>
<span class="text-[10px] {st['text_sub']}">Verified Analysis</span>
</div>
</div>
</div>
</div>
<div class="w-full md:w-[55%] overflow-y-auto custom-scroll p-6 md:p-10 flex flex-col gap-8 bg-inherit">
<div>
<div class="flex items-center gap-3 mb-5 border-b {divider_color} pb-3">
<div class="p-2 rounded-lg {st['tag_bg']}">
<svg class="w-6 h-6" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M8 10h.01M12 10h.01M16 10h.01M9 16H5a2 2 0 01-2-2V6a2 2 0 012-2h14a2 2 0 012 2v8a2 2 0 01-2 2h-5l-5 5v-5z"></path></svg>
</div>
<div>
<h2 class="text-xl md:text-2xl font-bold {st['text_main']}">私信回复话术</h2>
<p class="text-xs {st['text_sub']}">高情商回复，点击卡片即可复制</p>
</div>
</div>
<div class="space-y-1">
{thank_you_html}
</div>
</div>
<div>
<div class="flex items-center gap-3 mb-5 mt-2 border-b {divider_color} pb-3">
<div class="p-2 rounded-lg {st['tag_bg']}">
<svg class="w-6 h-6" fill="none" stroke="currentColor" viewBox="0 0 24 24"><path stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M12 8v13m0-13V6a2 2 0 112 2h-2zm0 0V5.5A2.5 2.5 0 109.5 8H12zm-7 4h14M5 12a2 2 0 110-4h14a2 2 0 110 4M5 12v7a2 2 0 002 2h10a2 2 0 002-2v-7"></path></svg>
</div>
<div>
<h2 class="text-xl md:text-2xl font-bold {st['text_main']}">推荐回礼策略</h2>
<p class="text-xs {st['text_sub']}">基于价格区间的最优解</p>
</div>
</div>
<div class="grid grid-cols-1 sm:grid-cols-2 gap-3">
{return_gift_html}
</div>
</div>
<div class="mt-auto pt-8 text-center opacity-40">
<p class="text-[10px] {st['text_sub']}">Designed by AI Gift Agent • 春节特别版</p>
</div>
</div>
</div>
<script>
function copyText(element, text) {{
navigator.clipboard.writeText(text).then(() => {{
const feedback = element.querySelector('.copy-feedback');
feedback.classList.remove('opacity-0');
feedback.classList.remove('pointer-events-none');
setTimeout(() => {{
feedback.classList.add('opacity-0');
feedback.classList.add('pointer-events-none');
}}, 1500);
}});
}}
</script>
</body>
</html>
"""
try:
directory = os.path.dirname(output_path)
if directory:
os.makedirs(directory, exist_ok=True)
with open(output_path, "w", encoding="utf-8") as f:
f.write(html_content)
return os.path.abspath(output_path)
except Exception as e:
return f"Error saving HTML file: {str(e)}"
if __name__ == "__main__":
parser = argparse.ArgumentParser(description="Generate Gift Card HTML")
parser.add_argument("action", nargs="?", help="Action command")
parser.add_argument("--product_name", required=True)
parser.add_argument("--price", required=True)
parser.add_argument("--evaluation", required=True)
parser.add_argument("--thank_you_json", required=True)
parser.add_argument("--return_gift_json", required=True)
parser.add_argument("--vibe_code", required=True)
parser.add_argument("--image_url", required=True)
parser.add_argument("--output_path", required=True)
args = parser.parse_args()
result_path = generate_gift_card(
product_name=args.product_name,
price=args.price,
evaluation=args.evaluation,
thank_you_json=args.thank_you_json,
return_gift_json=args.return_gift_json,
vibe_code=args.vibe_code,
image_url=args.image_url,
output_path=args.output_path
)
print(f"HTML Card generated successfully: {result_path}")