Skip to content

Commit 6645959

Browse files
pythonHuangclaude
andcommitted
docs: README更新AI智能编排体系(大模型设置多供应商/模型拉取测试+四个AI助手:流程编排/接口接入/报表/自定义助手多轮对话+设计器AI生成),中英文同步功能特性与v1.8更新日志
Co-Authored-By: Claude Code <noreply@anthropic.com>
1 parent 47f724a commit 6645959

1 file changed

Lines changed: 76 additions & 20 deletions

File tree

‎README.md‎

Lines changed: 76 additions & 20 deletions
Original file line numberDiff line numberDiff line change
@@ -63,16 +63,36 @@ docker-compose up -d
6363

6464
## 🤖 AI 智能编排
6565

66-
通过自然语言对话,让大模型自动生成接口流程编排,无需手动拖拽节点。
66+
通过自然语言对话,让大模型自动完成流程编排、接口接入与报表设计,无需手动操作。
6767

68-
### 使用方式
68+
### 大模型设置
6969

70-
1. 进入**流程设计器**,点击工具栏「AI 生成」按钮
71-
2. 展开「模型设置」,填写 OpenAI 兼容接口的地址、密钥与模型名(支持 DeepSeek / 通义千问 / Kimi / OpenAI 等,配置保存在系统配置中)
72-
3. 在需求描述中说明编排需求,例如:
73-
> 接收用户id,先调用获取用户信息接口,再根据用户id查询该用户的订单列表,最后返回用户名称和订单列表
74-
4. 点击「生成并替换画布」— AI 会从已加载的接口清单中选择接口、自动映射入参(流程入参)与出参(env_ 变量)、自动连线,并支持生成条件分支(CONDITION)、数据库节点(MYSQL)等 14 种节点
75-
5. 生成后可在画布上继续手工调整,保存/部署后即可对外提供接口
70+
系统设置 → 大模型设置:接入任意 **OpenAI 兼容接口**(DeepSeek / 通义千问 / Kimi / OpenAI / 智谱GLM / Ollama 等):
71+
72+
- 供应商预置下拉(可直接输入自定义),选择后自动填默认接口地址
73+
- 「获取模型/测试」一键验证密钥并拉取可用模型列表
74+
- 默认模型下拉选择、可用模型多选(对话中可切换供应商与模型)
75+
- 多供应商并存,支持启用/禁用
76+
77+
### 模型助手
78+
79+
菜单「模型助手」提供四个 AI 助手:
80+
81+
| 助手 | 能力 |
82+
|------|------|
83+
| **模型助手管理** | 自定义智能助手:名称、系统提示词、输入参数列表(文本/数字/日期/开关/下拉)、输出参数列表、辅助提问词按钮;启用后自动加入菜单 |
84+
| **流程智能编排助手** | 对话描述需求 → 自动选择已有接口、生成流程编排与**入参/出参定义**、条件分支、数据库节点等 14 种节点 → 预览确认后创建流程(参数随流程入库) |
85+
| **接口智能接入助手** | 对话描述接入需求 → 生成套件与接口方案(**入参/Header/出参配置**)→ 预览确认后写入平台(按 code 去重) |
86+
| **报表智能助手** | 对话描述报表需求 → 自动生成数据集(优先复用数据视图/数据源/流程/接口)、查询参数、排版(汇总用聚合、条件用公式)→ 预览确认后创建报表 |
87+
88+
自定义助手支持**多轮对话**:聊天界面气泡消息、完整历史上下文、结束对话时按输出参数生成最终 JSON 结果、开启新对话与历史记录查看。
89+
90+
### 流程设计器内 AI 生成
91+
92+
流程设计器工具栏同样提供「AI 生成」按钮,例如:
93+
> 接收用户id,先调用获取用户信息接口,再根据用户id查询该用户的订单列表,最后返回用户名称和订单列表
94+
95+
AI 自动选接口、映射入参(流程入参)与出参(env_ 变量)、连线并生成条件分支,确认后可直接在画布上继续调整,保存/部署即可对外提供接口。
7696

7797
### 支持的节点
7898

@@ -97,7 +117,6 @@ START / END / METHOD(接口调用)/ CONDITION(条件分支)/ MERGE(汇
97117
### 核心流程
98118

99119
- ✅ 可视化流程设计器(节点画布)
100-
- ✅ **🤖 AI 智能编排** — 对话式描述需求,大模型自动生成接口流程编排(OpenAI 兼容接口,支持 DeepSeek/通义/Kimi 等)
101120
- ✅ **14 种节点**:START / END / METHOD / CONDITION / MERGE / ASSIGN / CODE / DB / SUB_FLOW / LOOP / DELAY / PARALLEL / NOTIFY / TRANSFORM(模板转换)
102121
- ✅ **对象子属性级联选择**(赋值/方法节点支持 object/array 类型属性树形选择)
103122
- ✅ **数组操作**(赋值节点支持分页/取第n个/转JSON)
@@ -107,6 +126,15 @@ START / END / METHOD(接口调用)/ CONDITION(条件分支)/ MERGE(汇
107126
- ✅ 流程克隆 / 导入 / 导出(含 Word 文档)
108127
- ✅ 流程分组管理
109128

129+
### 模型助手(AI)
130+
131+
- ✅ **大模型设置** — 多供应商管理(预置下拉/模型拉取/配置测试/启停),支持任意 OpenAI 兼容接口(DeepSeek/通义千问/Kimi/OpenAI/智谱/Ollama)
132+
- ✅ **流程智能编排助手** — 对话生成流程(自动选接口/入参出参映射/条件分支/14 种节点),预览确认入库
133+
- ✅ **接口智能接入助手** — 对话生成套件与接口(含入参/Header/出参配置),预览确认接入
134+
- ✅ **报表智能助手** — 对话生成报表(复用数据视图/数据源/流程/接口 + 查询参数 + 排版公式),预览确认创建
135+
- ✅ **自定义智能助手** — 名称/系统提示词/输入输出参数/辅助提问词,启用后加入菜单;多轮对话 + 结束输出最终结果 + 历史记录
136+
- ✅ **流程设计器 AI 生成** — 工具栏一键生成并替换画布
137+
110138
### 触发方式
111139

112140
- ✅ 同步触发 `GET/POST /open/flow/trigger/{key}`
@@ -167,7 +195,7 @@ START / END / METHOD(接口调用)/ CONDITION(条件分支)/ MERGE(汇
167195

168196
### v1.8(最新)
169197

170-
- 🤖 **AI 智能编排** — 接入大模型能力(OpenAI 兼容接口,支持 DeepSeek/通义千问/Kimi 等),设计器对话式描述需求自动生成接口流程编排(自动选接口/映射入参出参/连线/条件分支,支持 14 种节点)
198+
- 🤖 **AI 智能编排体系** — 接入大模型能力(OpenAI 兼容接口,支持 DeepSeek/通义千问/Kimi/OpenAI/智谱/Ollama 等):大模型设置(多供应商/模型拉取/配置测试/启停);流程智能编排助手(对话生成流程+入参出参,预览确认);接口智能接入助手(对话生成套件接口+入参/Header/出参,预览确认);报表智能助手(对话生成报表:复用数据视图/数据源 + 查询参数 + 排版公式,预览确认);自定义智能助手(提示词/输入输出参数/辅助提问词,多轮对话+结束输出+历史记录);设计器 AI 生成
171199
- 📊 **报表模块** — 新增数据视图 + 报表设计器 + 公式引擎(SUM/AVG/COUNT/IF 等);数据集管理支持 4 种数据源与字段树点击填入;A4 分页预览、撤销/重做、单元格直接编辑、行列插入删除
172200
- 📈 **监控模块** — API 拓扑图(健康检查/访问统计/状态红黄绿/DB 节点与连线/流程过滤)+ 告警规则 + 告警记录
173201
- 🧮 **表达式引擎** — 条件/赋值节点支持算术运算 + - * / %、字符串拼接/切片([..5]/[2..5])/replace、toString("#.0##") 数字与日期格式化;赋值节点新增 EXPRESSION 表达式来源
@@ -294,16 +322,36 @@ docker-compose up -d
294322

295323
## 🤖 AI Orchestration
296324

297-
Generate API flow orchestration from natural language requirements — no manual node dragging needed.
325+
Use natural language to let the LLM auto-complete flow orchestration, API integration and report design.
298326

299-
### How to Use
327+
### Model Settings
300328

301-
1. Open the **Flow Designer** and click the "AI Generate" button in the toolbar
302-
2. Expand "Model Settings" and fill in an OpenAI-compatible API base URL, API key and model name (DeepSeek / Qwen / Kimi / OpenAI etc.; saved in system config)
303-
3. Describe your requirement, e.g.:
304-
> Take a userId, call the get-user-info API, then query the user's order list, finally return the user name and the order list
305-
4. Click "Generate & Replace Canvas" — the LLM picks APIs from the available list, auto-maps inputs (flow input params) and outputs (`env_` variables), wires nodes automatically, and supports all 14 node types including CONDITION branches and MYSQL (SQL) nodes
306-
5. Fine-tune the result on the canvas, then save/deploy to expose it as an API
329+
System Settings → AI Models: connect any **OpenAI-compatible API** (DeepSeek / Qwen / Kimi / OpenAI / GLM / Ollama etc.):
330+
331+
- Preset provider dropdown (or type your own) with auto-filled default base URL
332+
- "Fetch models / Test" verifies the API key and pulls the available model list
333+
- Default model dropdown + available models multi-select (switchable during chats)
334+
- Multiple providers with enable/disable
335+
336+
### AI Assistants
337+
338+
The "AI Assistants" menu provides four assistants:
339+
340+
| Assistant | Capability |
341+
|-----------|------------|
342+
| **Assistant Management** | Custom assistants: name, system prompt, input params (text/number/date/switch/select), output params, quick-prompt buttons; auto-added to menu when enabled |
343+
| **Flow Orchestration Assistant** | Describe requirements → auto-picks existing APIs, generates flow + input/output param definitions, condition branches, DB nodes (14 types) → preview, then confirm to create the flow |
344+
| **API Integration Assistant** | Describe integration needs → generates suites & APIs (input / Header / output param configs) → preview, then confirm to import (deduplicated by code) |
345+
| **Report Assistant** | Describe report needs → auto-generates datasets (reuses data views / data sources / flows / APIs), query params, layout (aggregates & formulas) → preview, then confirm to create the report |
346+
347+
Custom assistants support **multi-turn conversations**: chat bubbles, full history context, end-conversation to produce final JSON results from output params, new conversation and history viewing.
348+
349+
### In-Designer AI Generation
350+
351+
The Flow Designer toolbar also has an "AI Generate" button, e.g.:
352+
> Take a userId, call the get-user-info API, then query the user's order list, finally return the user name and the order list
353+
354+
The LLM picks APIs, auto-maps inputs (flow input params) and outputs (`env_` variables), wires nodes with condition branches; confirm to apply, then fine-tune on the canvas and save/deploy.
307355

308356
### Supported Node Types
309357

@@ -328,7 +376,6 @@ START / END / METHOD / CONDITION / MERGE / ASSIGN / CODE / MYSQL / LOOP / DELAY
328376
### Core Workflow
329377

330378
- ✅ Visual workflow designer (node canvas)
331-
- ✅ **🤖 AI orchestration** — Describe requirements in natural language and let the LLM auto-generate flow orchestration (OpenAI-compatible API; DeepSeek/Qwen/Kimi supported)
332379
- ✅ **14 Node Types**: START / END / METHOD / CONDITION / MERGE / ASSIGN / CODE / DB / SUB_FLOW / LOOP / DELAY / PARALLEL / NOTIFY / TRANSFORM
333380
- ✅ **Object sub-property cascading** (ASSIGN/METHOD nodes support object/array tree property selection)
334381
- ✅ **Array operations** (ASSIGN node: paginate / get by index / to JSON)
@@ -338,6 +385,15 @@ START / END / METHOD / CONDITION / MERGE / ASSIGN / CODE / MYSQL / LOOP / DELAY
338385
- ✅ Clone / Import / Export (with Word docs)
339386
- ✅ Workflow grouping
340387

388+
### AI Assistants
389+
390+
- ✅ **Model settings** — Multi-provider management (preset dropdown / model fetch / config test / enable-disable), any OpenAI-compatible API (DeepSeek/Qwen/Kimi/OpenAI/GLM/Ollama)
391+
- ✅ **Flow orchestration assistant** — Chat to generate flows (auto API selection, I/O param mapping, condition branches, 14 node types), preview then confirm
392+
- ✅ **API integration assistant** — Chat to generate suites & APIs (input / Header / output param configs), preview then confirm
393+
- ✅ **Report assistant** — Chat to generate reports (reuses data views / data sources / flows / APIs + query params + layout formulas), preview then confirm
394+
- ✅ **Custom assistants** — Name / system prompt / I/O params / quick prompts, auto-added to menu; multi-turn chat + final structured outputs + history
395+
- ✅ **In-designer AI generation** — One-click generate and replace the canvas
396+
341397
### Trigger Methods
342398

343399
- ✅ Synchronous trigger `GET/POST /open/flow/trigger/{key}`
@@ -386,7 +442,7 @@ START / END / METHOD / CONDITION / MERGE / ASSIGN / CODE / MYSQL / LOOP / DELAY
386442

387443
### v1.8(Latest)
388444

389-
- 🤖 **AI orchestration** — LLM integration (OpenAI-compatible API; DeepSeek/Qwen/Kimi supported); describe requirements in the designer to auto-generate flow orchestration (auto API selection, input/output mapping, wiring, condition branches; all 14 node types)
445+
- 🤖 **AI orchestration suite** — LLM integration (OpenAI-compatible API; DeepSeek/Qwen/Kimi/OpenAI/GLM/Ollama supported): model settings (multi-provider / model fetch / config test / enable-disable); flow orchestration assistant (chat to generate flows with I/O params, preview then confirm); API integration assistant (chat to generate suites & APIs with input/Header/output configs, preview then confirm); report assistant (chat to generate reports reusing data views/data sources + query params + layout formulas, preview then confirm); custom assistants (prompts / I/O params / quick prompts, multi-turn chat + final outputs + history); in-designer AI generation
390446
- 📊 **Report module** — Data views + Report designer + Formula engine (SUM/AVG/COUNT/IF etc.); datasets with 4 source types and field-tree click-to-fill; A4 paged preview, undo/redo, inline cell editing, row/column insert & delete
391447
- 📈 **Monitoring module** — API topology map (health check / visit stats / red-yellow-green status / DB nodes & edges / flow filter) + alert rules + alert records
392448
- 🧮 **Expression engine** — CONDITION/ASSIGN nodes support arithmetic + - * / %, string concat / slice ([..5]/[2..5]) / replace, toString("#.0##") number & date formatting; ASSIGN node adds EXPRESSION source type

0 commit comments

Comments
 (0)