@aigne/cli 1.17.0 → 1.18.0

package/CHANGELOG.md

@@ -1,5 +1,28 @@
 # Changelog
 
+## [1.18.0](https://github.com/AIGNE-io/aigne-framework/compare/cli-v1.17.0...cli-v1.18.0) (2025-07-01)
+
+
+### Features
+
+* **example:** use AIGNE cli to run chat-bot example ([#198](https://github.com/AIGNE-io/aigne-framework/issues/198)) ([7085541](https://github.com/AIGNE-io/aigne-framework/commit/708554100692f2a557f7329ea78e46c3c870ce10))
+
+
+### Dependencies
+
+* The following workspace dependencies were updated
+  * dependencies
+    * @aigne/agent-library bumped to 1.17.5
+    * @aigne/anthropic bumped to 0.5.0
+    * @aigne/bedrock bumped to 0.5.0
+    * @aigne/core bumped to 1.28.0
+    * @aigne/deepseek bumped to 0.4.0
+    * @aigne/gemini bumped to 0.4.0
+    * @aigne/ollama bumped to 0.4.0
+    * @aigne/open-router bumped to 0.4.0
+    * @aigne/openai bumped to 0.6.0
+    * @aigne/xai bumped to 0.4.0
+
 ## [1.17.0](https://github.com/AIGNE-io/aigne-framework/compare/cli-v1.16.0...cli-v1.17.0) (2025-07-01)
 
 

package/README.md

@@ -16,12 +16,12 @@ Command-line tool for [AIGNE Framework](https://github.com/AIGNE-io/aigne-framew
 
 ## Features
 
-- **Project Creation**: Quickly create new AIGNE projects with predefined file structures and configurations
-- **Agent Running**: Easily run and test AIGNE agents
-- **Testing Support**: Built-in test command for unit testing and integration testing
-- **MCP Services**: Support for launching agents as MCP servers for integration with external systems
-- **Interactive Interface**: Beautiful command-line interface providing an intuitive user experience
-- **Multi-model Support**: Support for OpenAI, Claude, XAI, and other model providers
+* **Project Creation**: Quickly create new AIGNE projects with predefined file structures and configurations
+* **Agent Running**: Easily run and test AIGNE agents
+* **Testing Support**: Built-in test command for unit testing and integration testing
+* **MCP Services**: Support for launching agents as MCP servers for integration with external systems
+* **Interactive Interface**: Beautiful command-line interface providing an intuitive user experience
+* **Multi-model Support**: Support for OpenAI, Claude, XAI, and other model providers
 
 ## Installation
 
@@ -81,8 +81,8 @@ aigne create my-project
 
 The interactive creation process will ask for:
 
-- Project name
-- Project template (currently supports the default template)
+* Project name
+* Project template (currently supports the default template)
 
 ## Run Command
 
@@ -104,10 +104,10 @@ aigne run --entry-agent myAgent
 
 Available options:
 
-- `--entry-agent <entry-agent>` - Specify the agent name to run (defaults to the first agent found)
-- `--cache-dir <dir>` - Specify the directory to download the package to (used in URL mode)
-- `--model <provider[:model]>` - Specify the AI model in format 'provider\[:model]' where model is optional (e.g., 'openai' or 'openai:gpt-4o-mini')
-- `--verbose` - Enable verbose logging
+* `--entry-agent <entry-agent>` - Specify the agent name to run (defaults to the first agent found)
+* `--cache-dir <dir>` - Specify the directory to download the package to (used in URL mode)
+* `--model <provider[:model]>` - Specify the AI model in format 'provider\[:model]' where model is optional (e.g., 'openai' or 'openai:gpt-4o-mini')
+* `--verbose` - Enable verbose logging
 
 ## Test Command
 

package/README.zh.md

@@ -16,12 +16,12 @@
 
 ## 特性
 
-- **项目创建**：快速创建新的 AIGNE 项目，包含预设的文件结构和配置
-- **代理运行**：轻松运行和测试 AIGNE 代理
-- **测试支持**：内置测试命令，方便进行单元测试和集成测试
-- **MCP 服务**：支持将代理作为 MCP 服务器启动，与外部系统集成
-- **交互式界面**：美观的命令行界面，提供直观的使用体验
-- **多模型支持**：支持 OpenAI、Claude、XAI 等多种模型提供商
+* **项目创建**：快速创建新的 AIGNE 项目，包含预设的文件结构和配置
+* **代理运行**：轻松运行和测试 AIGNE 代理
+* **测试支持**：内置测试命令，方便进行单元测试和集成测试
+* **MCP 服务**：支持将代理作为 MCP 服务器启动，与外部系统集成
+* **交互式界面**：美观的命令行界面，提供直观的使用体验
+* **多模型支持**：支持 OpenAI、Claude、XAI 等多种模型提供商
 
 ## 安装
 
@@ -81,8 +81,8 @@ aigne create my-project
 
 交互式创建过程会询问：
 
-- 项目名称
-- 项目模板（目前支持 default 模板）
+* 项目名称
+* 项目模板（目前支持 default 模板）
 
 ## 运行命令 (run)
 
@@ -104,10 +104,10 @@ aigne run --entry-agent myAgent
 
 可用选项：
 
-- `--entry-agent <代理>` - 指定要运行的代理名称（默认为找到的第一个代理）
-- `--cache-dir <目录>` - 指定下载包的目录（URL模式下使用）
-- `--model <提供商[:模型]>` - 指定AI模型，格式为'提供商\[:模型]'，其中模型是可选的（如'openai'或'openai:gpt-4o-mini'）
-- `--verbose` - 启用详细日志记录
+* `--entry-agent <代理>` - 指定要运行的代理名称（默认为找到的第一个代理）
+* `--cache-dir <目录>` - 指定下载包的目录（URL模式下使用）
+* `--model <提供商[:模型]>` - 指定AI模型，格式为'提供商\[:模型]'，其中模型是可选的（如'openai'或'openai:gpt-4o-mini'）
+* `--verbose` - 启用详细日志记录
 
 ## 测试命令 (test)
 

package/dist/commands/aigne.js

@@ -2,7 +2,7 @@ import { Command } from "commander";
 import { AIGNE_CLI_VERSION } from "../constants.js";
 import { asciiLogo } from "../utils/ascii-logo.js";
 import { createCreateCommand } from "./create.js";
-import { createObservabilityCommand } from "./observability.js";
+import { createObservabilityCommand } from "./observe.js";
 import { createRunCommand } from "./run.js";
 import { createServeCommand } from "./serve.js";
 import { createTestCommand } from "./test.js";

package/dist/commands/{observability.js → observe.js}

@@ -1,6 +1,6 @@
 import { tryOrThrow } from "@aigne/core/utils/type-utils.js";
-import { startObservabilityCLIServer } from "@aigne/observability/cli";
-import getObservabilityDbPath from "@aigne/observability/db-path";
+import { startObservabilityCLIServer } from "@aigne/observability-api/cli";
+import getObservabilityDbPath from "@aigne/observability-api/db-path";
 import { Command } from "commander";
 import detectPort from "detect-port";
 const DEFAULT_PORT = () => tryOrThrow(() => {
@@ -20,7 +20,7 @@ export function createObservabilityCommand() {
         .action(async (options) => {
         const port = await detectPort(options.port || DEFAULT_PORT());
         const dbUrl = getObservabilityDbPath();
-        console.log("DB PATH:", dbUrl);
+        console.log("Observability database path:", dbUrl);
         await startObservabilityCLIServer({ port, dbUrl });
     })
         .showHelpAfterError(true)

package/dist/tracer/terminal.js

@@ -148,7 +148,7 @@ export class TerminalTracer {
     }
     formatResult(agent, context, m = {}) {
         const { isTTY } = process.stdout;
-        const outputKey = agent instanceof AIAgent ? agent.outputKey : undefined;
+        const outputKey = this.outputKey || (agent instanceof AIAgent ? agent.outputKey : undefined);
         const prefix = logger.enabled(LogLevel.INFO)
             ? `${chalk.grey(figures.tick)} 🤖 ${this.formatTokenUsage(context.usage)}`
             : null;

package/dist/utils/run-chat-loop.d.ts

@@ -6,5 +6,6 @@ export interface ChatLoopOptions {
     defaultQuestion?: string;
     inputKey?: string;
     skipLoop?: boolean;
+    outputKey?: string;
 }
 export declare function runChatLoopInTerminal(userAgent: UserAgent<any, any>, options?: ChatLoopOptions): Promise<void>;

package/dist/utils/run-chat-loop.js

@@ -42,7 +42,7 @@ export async function runChatLoopInTerminal(userAgent, options = {}) {
     }
 }
 async function callAgent(userAgent, input, options) {
-    const tracer = new TerminalTracer(userAgent.context);
+    const tracer = new TerminalTracer(userAgent.context, options);
     await tracer.run(userAgent, typeof input === "string" ? { [options.inputKey || DEFAULT_CHAT_INPUT_KEY]: input } : input);
 }
 const COMMANDS = {

package/dist/utils/run-with-aigne.d.ts

@@ -13,6 +13,7 @@ export interface RunAIGNECommandOptions {
     input?: string[];
     format?: "text" | "json" | "yaml";
     output?: string;
+    outputKey?: string;
     logLevel?: LogLevel;
     force?: boolean;
 }

package/dist/utils/run-with-aigne.js

@@ -124,7 +124,7 @@ export async function runWithAIGNE(agentCreator, { argv = process.argv, chatLoop
             }
             await runAgentWithAIGNE(aigne, agent, {
                 ...options,
-                outputKey,
+                outputKey: outputKey || options.outputKey,
                 chatLoopOptions,
                 modelOptions,
                 input,
@@ -168,6 +168,7 @@ export async function runAgentWithAIGNE(aigne, agent, { outputKey, chatLoopOptio
         const userAgent = agent instanceof UserAgent ? agent : aigne.invoke(agent);
         await runChatLoopInTerminal(userAgent, {
             ...chatLoopOptions,
+            outputKey,
         });
         return;
     }

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@aigne/cli",
-  "version": "1.17.0",
+  "version": "1.18.0",
   "description": "cli for AIGNE framework",
   "publishConfig": {
     "access": "public"
@@ -55,17 +55,17 @@
     "wrap-ansi": "^9.0.0",
     "yaml": "^2.7.1",
     "zod": "^3.24.4",
-    "@aigne/agent-library": "^1.17.4",
-    "@aigne/bedrock": "^0.4.0",
-    "@aigne/anthropic": "^0.4.0",
-    "@aigne/core": "^1.27.0",
-    "@aigne/gemini": "^0.3.11",
-    "@aigne/ollama": "^0.3.11",
-    "@aigne/observability": "^0.3.0",
-    "@aigne/deepseek": "^0.3.11",
-    "@aigne/open-router": "^0.3.11",
-    "@aigne/openai": "^0.5.0",
-    "@aigne/xai": "^0.3.11"
+    "@aigne/agent-library": "^1.17.5",
+    "@aigne/anthropic": "^0.5.0",
+    "@aigne/core": "^1.28.0",
+    "@aigne/deepseek": "^0.4.0",
+    "@aigne/gemini": "^0.4.0",
+    "@aigne/bedrock": "^0.5.0",
+    "@aigne/observability-api": "^0.3.0",
+    "@aigne/open-router": "^0.4.0",
+    "@aigne/ollama": "^0.4.0",
+    "@aigne/openai": "^0.6.0",
+    "@aigne/xai": "^0.4.0"
   },
   "devDependencies": {
     "@types/archiver": "^6.0.3",

package/templates/default/.env.local.example

@@ -1,18 +1,44 @@
 # Change the name of this file to .env.local and fill in the following values
 
-# Use this for OpenAI models
+# Uncomment the lines below to enable debug logging
+# DEBUG="aigne:*"
+
+# Use different Models
+
+# OpenAI
 MODEL="openai:gpt-4.1"
-OPENAI_API_KEY=""
+OPENAI_API_KEY="YOUR_OPENAI_API_KEY"
 
-# Use this for Anthropic models
+# Anthropic claude
 # MODEL="anthropic:claude-3-7-sonnet-latest"
-# ANTHROPIC_API_KEY="" # Your Anthropic API key
+# ANTHROPIC_API_KEY=""
 
-# Use this for AWS Bedrock models
+# Gemini
+# MODEL="gemini:gemini-2.0-flash"
+# GEMINI_API_KEY=""
+
+# Bedrock nova
+# MODEL=bedrock:us.amazon.nova-premier-v1:0
 # AWS_ACCESS_KEY_ID=""
 # AWS_SECRET_ACCESS_KEY=""
 # AWS_REGION=us-west-2
-# MODEL=Bedrock:us.amazon.nova-premier-v1:0
+
+# DeepSeek
+# MODEL="deepseek:deepseek-chat"
+# DEEPSEEK_API_KEY=""
+
+# OpenRouter
+# MODEL="openrouter:openai/gpt-4o"
+# OPEN_ROUTER_API_KEY=""
+
+# xAI
+# MODEL="xai:grok-2-latest"
+# XAI_API_KEY=""
+
+# Ollama
+# MODEL="ollama:llama3.2"
+# OLLAMA_DEFAULT_BASE_URL="http://localhost:11434/v1";
+
 
 # Setup proxy if needed
 # HTTPS_PROXY=http://localhost:7890

package/templates/default/README.md

@@ -17,6 +17,16 @@ This is the default project template for the AIGNE framework, providing a basic
 npm install -g aigne
 ```
 
+### Setup Environment Variables
+
+Copy the `.env.local.example` file to `.env.local` and set your OpenAI API key:
+
+```shell
+# OpenAI
+MODEL="openai:gpt-4.1"
+OPENAI_API_KEY="YOUR_OPENAI_API_KEY"
+```
+
 ### Start the Project
 
 ```bash