opc-agent 1.3.2 → 1.4.0

package/templates/docker-compose.yml

@@ -1,21 +1,21 @@
-version: '3.8'
-
-services:
-  agent:
-    build: .
-    ports:
-      - "3000:3000"
-    env_file:
-      - .env
-    volumes:
-      - ./oad.yaml:/app/oad.yaml:ro
-      - agent-data:/app/data
-    restart: unless-stopped
-    healthcheck:
-      test: ["CMD", "wget", "-q", "--spider", "http://localhost:3000/health"]
-      interval: 30s
-      timeout: 5s
-      retries: 3
-
-volumes:
-  agent-data:
+version: '3.8'
+
+services:
+  agent:
+    build: .
+    ports:
+      - "3000:3000"
+    env_file:
+      - .env
+    volumes:
+      - ./oad.yaml:/app/oad.yaml:ro
+      - agent-data:/app/data
+    restart: unless-stopped
+    healthcheck:
+      test: ["CMD", "wget", "-q", "--spider", "http://localhost:3000/health"]
+      interval: 30s
+      timeout: 5s
+      retries: 3
+
+volumes:
+  agent-data:

package/templates/ecommerce-assistant/README.md

@@ -1,45 +1,45 @@
-# Ecommerce Assistant — 电商助手工位
-
-专为电商平台设计的 AI 导购与售后工位，覆盖商品推荐、订单查询、退换货处理全链路。
-
-## 快速开始
-
-```bash
-opc init --template ecommerce-assistant
-opc start
-```
-
-访问 http://localhost:3000 即可使用电商助手聊天界面。
-
-## 功能
-
-| 技能 | 说明 |
-|------|------|
-| product-search | 关键词/类目/价格区间商品搜索 |
-| order-query | 订单状态与物流实时查询 |
-| after-sale | 退换货受理与投诉处理 |
-| promotion | 个性化优惠券与促销推送 |
-| recommendation | 基于偏好的个性化商品推荐 |
-
-## 配置
-
-在 `oad.yaml` 中修改以下参数：
-
-- `spec.provider.default` — 切换 LLM 提供商（deepseek / openai / qwen）
-- `spec.model` — 指定模型版本
-- `spec.systemPrompt` — 定制品牌话术、商品范围和售后规则
-
-### 环境变量
-
-```bash
-OPC_LLM_API_KEY=your_key
-OPC_LLM_MODEL=deepseek-chat        # 可选，覆盖 oad.yaml 中的 model
-OPC_DEEPBRAIN_ENABLED=true         # 启用 DeepBrain 知识库增强（需全局安装 deepbrain）
-```
-
-## 推荐搭配
-
-- 将商品目录、售后政策、FAQ 上传至知识库（`/api/kb/upload`）
-- 开启 `OPC_DEEPBRAIN_ENABLED=true` 提升商品语义匹配精度
-- 通过 Dashboard（`/dashboard`）监控转化率、客满率和退单率
-- 结合 CRM 系统 Webhook 实现订单状态实时同步
+# Ecommerce Assistant — 电商助手工位
+
+专为电商平台设计的 AI 导购与售后工位，覆盖商品推荐、订单查询、退换货处理全链路。
+
+## 快速开始
+
+```bash
+opc init --template ecommerce-assistant
+opc start
+```
+
+访问 http://localhost:3000 即可使用电商助手聊天界面。
+
+## 功能
+
+| 技能 | 说明 |
+|------|------|
+| product-search | 关键词/类目/价格区间商品搜索 |
+| order-query | 订单状态与物流实时查询 |
+| after-sale | 退换货受理与投诉处理 |
+| promotion | 个性化优惠券与促销推送 |
+| recommendation | 基于偏好的个性化商品推荐 |
+
+## 配置
+
+在 `oad.yaml` 中修改以下参数：
+
+- `spec.provider.default` — 切换 LLM 提供商（deepseek / openai / qwen）
+- `spec.model` — 指定模型版本
+- `spec.systemPrompt` — 定制品牌话术、商品范围和售后规则
+
+### 环境变量
+
+```bash
+OPC_LLM_API_KEY=your_key
+OPC_LLM_MODEL=deepseek-chat        # 可选，覆盖 oad.yaml 中的 model
+OPC_DEEPBRAIN_ENABLED=true         # 启用 DeepBrain 知识库增强（需全局安装 deepbrain）
+```
+
+## 推荐搭配
+
+- 将商品目录、售后政策、FAQ 上传至知识库（`/api/kb/upload`）
+- 开启 `OPC_DEEPBRAIN_ENABLED=true` 提升商品语义匹配精度
+- 通过 Dashboard（`/dashboard`）监控转化率、客满率和退单率
+- 结合 CRM 系统 Webhook 实现订单状态实时同步

package/templates/ecommerce-assistant/oad.yaml

@@ -1,47 +1,47 @@
-apiVersion: opc/v1
-kind: Agent
-metadata:
-  name: ecommerce-assistant
-  version: 1.0.0
-  description: "电商助手工位 - 商品推荐、订单查询、售后服务一体化"
-  author: Deepleaper
-  license: Apache-2.0
-  marketplace:
-    certified: false
-    category: ecommerce
-spec:
-  provider:
-    default: deepseek
-    allowed: [openai, deepseek, qwen, gemini]
-  model: deepseek-chat
-  systemPrompt: |
-    你是一名专业的电商购物助手。
-    你帮助用户查询商品信息、比较价格、处理订单状态和售后问题。
-    根据用户需求推荐合适的商品，推送当前促销活动，并协助用户做出购买决策。
-    处理退换货申请时，需先核实订单信息，再按平台规则指引用户操作。
-    始终保持友好、热情、专业的服务态度。
-  skills:
-    - name: product-search
-      description: "搜索商品信息，支持关键词、类目、价格区间筛选"
-    - name: order-query
-      description: "查询订单状态、物流轨迹和预计送达时间"
-    - name: after-sale
-      description: "处理退换货申请、投诉受理和赔偿协商"
-    - name: promotion
-      description: "推送个性化优惠券、限时活动和会员权益"
-    - name: recommendation
-      description: "基于用户偏好和购买历史做商品推荐"
-  channels:
-    - type: web
-      port: 3000
-  memory:
-    shortTerm: true
-    longTerm: true
-  knowledge:
-    enabled: true
-    deepbrain: auto
-  dtv:
-    trust:
-      level: sandbox
-    value:
-      metrics: [conversion_rate, customer_satisfaction, avg_order_value, refund_rate]
+apiVersion: opc/v1
+kind: Agent
+metadata:
+  name: ecommerce-assistant
+  version: 1.0.0
+  description: "电商助手工位 - 商品推荐、订单查询、售后服务一体化"
+  author: Deepleaper
+  license: Apache-2.0
+  marketplace:
+    certified: false
+    category: ecommerce
+spec:
+  provider:
+    default: deepseek
+    allowed: [openai, deepseek, qwen, gemini]
+  model: deepseek-chat
+  systemPrompt: |
+    你是一名专业的电商购物助手。
+    你帮助用户查询商品信息、比较价格、处理订单状态和售后问题。
+    根据用户需求推荐合适的商品，推送当前促销活动，并协助用户做出购买决策。
+    处理退换货申请时，需先核实订单信息，再按平台规则指引用户操作。
+    始终保持友好、热情、专业的服务态度。
+  skills:
+    - name: product-search
+      description: "搜索商品信息，支持关键词、类目、价格区间筛选"
+    - name: order-query
+      description: "查询订单状态、物流轨迹和预计送达时间"
+    - name: after-sale
+      description: "处理退换货申请、投诉受理和赔偿协商"
+    - name: promotion
+      description: "推送个性化优惠券、限时活动和会员权益"
+    - name: recommendation
+      description: "基于用户偏好和购买历史做商品推荐"
+  channels:
+    - type: web
+      port: 3000
+  memory:
+    shortTerm: true
+    longTerm: true
+  knowledge:
+    enabled: true
+    deepbrain: auto
+  dtv:
+    trust:
+      level: sandbox
+    value:
+      metrics: [conversion_rate, customer_satisfaction, avg_order_value, refund_rate]

package/templates/knowledge-base/README.md

@@ -1,28 +1,28 @@
-# Knowledge Base Template
-
-RAG-powered knowledge base agent that answers questions from your company docs.
-
-## Features
-- **Semantic Search**: Uses DeepBrain for intelligent document retrieval
-- **Document Ingestion**: Add docs to build your knowledge base
-- **Source Citations**: Cites sources in answers
-
-## Quick Start
-```bash
-opc init my-kb --template knowledge-base
-cd my-kb
-# Optional: install deepbrain for full RAG support
-npm install deepbrain
-opc run
-```
-
-## How It Works
-1. Ingest company documents into DeepBrain
-2. User asks a question
-3. Agent searches for relevant document chunks
-4. LLM generates answer based on retrieved context
-
-## Configuration
-- `memory.longTerm.provider: deepbrain` — enables semantic memory
-- `memory.longTerm.collection` — DeepBrain collection name
-- Falls back to in-memory if deepbrain is not installed
+# Knowledge Base Template
+
+RAG-powered knowledge base agent that answers questions from your company docs.
+
+## Features
+- **Semantic Search**: Uses DeepBrain for intelligent document retrieval
+- **Document Ingestion**: Add docs to build your knowledge base
+- **Source Citations**: Cites sources in answers
+
+## Quick Start
+```bash
+opc init my-kb --template knowledge-base
+cd my-kb
+# Optional: install deepbrain for full RAG support
+npm install deepbrain
+opc run
+```
+
+## How It Works
+1. Ingest company documents into DeepBrain
+2. User asks a question
+3. Agent searches for relevant document chunks
+4. LLM generates answer based on retrieved context
+
+## Configuration
+- `memory.longTerm.provider: deepbrain` — enables semantic memory
+- `memory.longTerm.collection` — DeepBrain collection name
+- Falls back to in-memory if deepbrain is not installed

package/templates/knowledge-base/oad.yaml

@@ -1,38 +1,38 @@
-apiVersion: opc/v1
-kind: Agent
-metadata:
-  name: knowledge-base
-  version: 1.0.0
-  description: "RAG-powered knowledge base agent using DeepBrain for semantic search"
-  author: Deepleaper
-  license: Apache-2.0
-  marketplace:
-    certified: false
-    category: knowledge
-spec:
-  provider:
-    default: deepseek
-    allowed: [openai, deepseek, qwen]
-  model: deepseek-chat
-  systemPrompt: |
-    You are a knowledge base assistant. Answer questions using the company documents
-    and knowledge provided to you. If you don't have enough information, say so honestly.
-    Always cite sources when possible. Be accurate and concise.
-  skills:
-    - name: doc-search
-      description: "Search company documents using semantic search"
-    - name: doc-ingest
-      description: "Ingest new documents into the knowledge base"
-  channels:
-    - type: web
-      port: 3000
-  memory:
-    shortTerm: true
-    longTerm:
-      provider: deepbrain
-      collection: company-knowledge
-  dtv:
-    trust:
-      level: sandbox
-    value:
-      metrics: [queries_answered, accuracy_score, docs_indexed]
+apiVersion: opc/v1
+kind: Agent
+metadata:
+  name: knowledge-base
+  version: 1.0.0
+  description: "RAG-powered knowledge base agent using DeepBrain for semantic search"
+  author: Deepleaper
+  license: Apache-2.0
+  marketplace:
+    certified: false
+    category: knowledge
+spec:
+  provider:
+    default: deepseek
+    allowed: [openai, deepseek, qwen]
+  model: deepseek-chat
+  systemPrompt: |
+    You are a knowledge base assistant. Answer questions using the company documents
+    and knowledge provided to you. If you don't have enough information, say so honestly.
+    Always cite sources when possible. Be accurate and concise.
+  skills:
+    - name: doc-search
+      description: "Search company documents using semantic search"
+    - name: doc-ingest
+      description: "Ingest new documents into the knowledge base"
+  channels:
+    - type: web
+      port: 3000
+  memory:
+    shortTerm: true
+    longTerm:
+      provider: deepbrain
+      collection: company-knowledge
+  dtv:
+    trust:
+      level: sandbox
+    value:
+      metrics: [queries_answered, accuracy_score, docs_indexed]

package/templates/sales-assistant/README.md

@@ -1,26 +1,26 @@
-# Sales Assistant Template
-
-AI-powered sales assistant for product Q&A, lead capture, and appointment booking.
-
-## Features
-- **Product Q&A**: Answers questions about your products/services
-- **Lead Capture**: Collects prospect information (name, email, company)
-- **Appointment Booking**: Schedules demos and meetings
-
-## Quick Start
-```bash
-opc init my-sales-bot --template sales-assistant
-cd my-sales-bot
-opc run
-```
-
-## Configuration
-Edit `oad.yaml` to customize:
-- System prompt with your product details
-- Add your product FAQ data
-- Configure appointment scheduling rules
-
-## Value Metrics
-- `leads_captured` — Number of leads collected
-- `appointments_booked` — Demos/meetings scheduled
-- `conversion_rate` — Lead to appointment ratio
+# Sales Assistant Template
+
+AI-powered sales assistant for product Q&A, lead capture, and appointment booking.
+
+## Features
+- **Product Q&A**: Answers questions about your products/services
+- **Lead Capture**: Collects prospect information (name, email, company)
+- **Appointment Booking**: Schedules demos and meetings
+
+## Quick Start
+```bash
+opc init my-sales-bot --template sales-assistant
+cd my-sales-bot
+opc run
+```
+
+## Configuration
+Edit `oad.yaml` to customize:
+- System prompt with your product details
+- Add your product FAQ data
+- Configure appointment scheduling rules
+
+## Value Metrics
+- `leads_captured` — Number of leads collected
+- `appointments_booked` — Demos/meetings scheduled
+- `conversion_rate` — Lead to appointment ratio

package/templates/sales-assistant/oad.yaml

@@ -1,43 +1,43 @@
-apiVersion: opc/v1
-kind: Agent
-metadata:
-  name: sales-assistant
-  version: 1.0.0
-  description: "AI sales assistant with product Q&A, lead capture, and appointment booking"
-  author: Deepleaper
-  license: Apache-2.0
-  marketplace:
-    certified: false
-    category: sales
-spec:
-  provider:
-    default: deepseek
-    allowed: [openai, deepseek, qwen]
-  model: deepseek-chat
-  systemPrompt: |
-    You are a professional sales assistant. Your goals:
-    1. Answer product questions accurately and enthusiastically
-    2. Capture leads by collecting name, email, and company info
-    3. Book appointments when prospects are ready
-    Be friendly, persuasive but not pushy. Always provide value first.
-  skills:
-    - name: product-qa
-      description: "Answer product-related questions"
-    - name: lead-capture
-      description: "Capture prospect information"
-    - name: appointment-booking
-      description: "Schedule meetings and demos"
-  channels:
-    - type: web
-      port: 3000
-    - type: websocket
-      port: 3002
-  memory:
-    shortTerm: true
-    longTerm:
-      provider: in-memory
-  dtv:
-    trust:
-      level: sandbox
-    value:
-      metrics: [leads_captured, appointments_booked, conversion_rate]
+apiVersion: opc/v1
+kind: Agent
+metadata:
+  name: sales-assistant
+  version: 1.0.0
+  description: "AI sales assistant with product Q&A, lead capture, and appointment booking"
+  author: Deepleaper
+  license: Apache-2.0
+  marketplace:
+    certified: false
+    category: sales
+spec:
+  provider:
+    default: deepseek
+    allowed: [openai, deepseek, qwen]
+  model: deepseek-chat
+  systemPrompt: |
+    You are a professional sales assistant. Your goals:
+    1. Answer product questions accurately and enthusiastically
+    2. Capture leads by collecting name, email, and company info
+    3. Book appointments when prospects are ready
+    Be friendly, persuasive but not pushy. Always provide value first.
+  skills:
+    - name: product-qa
+      description: "Answer product-related questions"
+    - name: lead-capture
+      description: "Capture prospect information"
+    - name: appointment-booking
+      description: "Schedule meetings and demos"
+  channels:
+    - type: web
+      port: 3000
+    - type: websocket
+      port: 3002
+  memory:
+    shortTerm: true
+    longTerm:
+      provider: in-memory
+  dtv:
+    trust:
+      level: sandbox
+    value:
+      metrics: [leads_captured, appointments_booked, conversion_rate]

package/templates/tech-support/README.md

@@ -1,43 +1,43 @@
-# Tech Support Agent — 技术支持工位
-
-专为 IT 支持团队设计的 AI 工位，覆盖软件故障、系统配置、网络问题等场景。
-
-## 快速开始
-
-```bash
-opc init --template tech-support
-opc start
-```
-
-访问 http://localhost:3000 即可使用技术支持聊天界面。
-
-## 功能
-
-| 技能 | 说明 |
-|------|------|
-| troubleshoot | 分步骤诊断和解决技术问题 |
-| knowledge-lookup | 查询技术文档与历史解决方案 |
-| ticket-create | 创建并跟踪技术支持工单 |
-| escalate | 升级复杂问题至专项团队 |
-
-## 配置
-
-在 `oad.yaml` 中修改以下参数：
-
-- `spec.provider.default` — 切换 LLM 提供商（deepseek / openai / qwen）
-- `spec.model` — 指定模型版本
-- `spec.systemPrompt` — 定制支持范围和话术风格
-
-### 环境变量
-
-```bash
-OPC_LLM_API_KEY=your_key
-OPC_LLM_MODEL=deepseek-chat        # 可选，覆盖 oad.yaml 中的 model
-OPC_DEEPBRAIN_ENABLED=true         # 启用 DeepBrain 知识库增强（需全局安装 deepbrain）
-```
-
-## 推荐搭配
-
-- 将内部技术文档、SOP、FAQ 上传至知识库（`/api/kb/upload`）
-- 开启 `OPC_DEEPBRAIN_ENABLED=true` 获得更精准的语义检索
-- 通过 Dashboard（`/dashboard`）监控首次解决率和平均响应时间
+# Tech Support Agent — 技术支持工位
+
+专为 IT 支持团队设计的 AI 工位，覆盖软件故障、系统配置、网络问题等场景。
+
+## 快速开始
+
+```bash
+opc init --template tech-support
+opc start
+```
+
+访问 http://localhost:3000 即可使用技术支持聊天界面。
+
+## 功能
+
+| 技能 | 说明 |
+|------|------|
+| troubleshoot | 分步骤诊断和解决技术问题 |
+| knowledge-lookup | 查询技术文档与历史解决方案 |
+| ticket-create | 创建并跟踪技术支持工单 |
+| escalate | 升级复杂问题至专项团队 |
+
+## 配置
+
+在 `oad.yaml` 中修改以下参数：
+
+- `spec.provider.default` — 切换 LLM 提供商（deepseek / openai / qwen）
+- `spec.model` — 指定模型版本
+- `spec.systemPrompt` — 定制支持范围和话术风格
+
+### 环境变量
+
+```bash
+OPC_LLM_API_KEY=your_key
+OPC_LLM_MODEL=deepseek-chat        # 可选，覆盖 oad.yaml 中的 model
+OPC_DEEPBRAIN_ENABLED=true         # 启用 DeepBrain 知识库增强（需全局安装 deepbrain）
+```
+
+## 推荐搭配
+
+- 将内部技术文档、SOP、FAQ 上传至知识库（`/api/kb/upload`）
+- 开启 `OPC_DEEPBRAIN_ENABLED=true` 获得更精准的语义检索
+- 通过 Dashboard（`/dashboard`）监控首次解决率和平均响应时间