开始使用pi coding agent(通过阿里云的Dashscope配置Kimi-K3)

先决条件

1.有阿里云Dashscope的API KEY，没有的话可以去申请
2.本地安装pnpm，有基础的nodejs和pnpm包管理环境

pi coding agent的安装与升级

通过pnpm安装：

pnpm add -g --ignore-scripts @earendil-works/pi-coding-agent

出现如下提示表示成功：

[WARN] 1 deprecated subdependencies found: node-domexception@1.0.0
Packages: +128
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Progress: resolved 137, reused 9, downloaded 119, added 120, done

global:
+ @earendil-works/pi-coding-agent 0.82.1 (0.83.0 is available)

Done in 10.6s using pnpm v11.8.0

通过pnpm升级（搞不懂为什么刚安装就需要升级，我看了下版本确实升级了）：

pi update

出现如下提示表示成功：

Updating pi with pnpm install -g --ignore-scripts --config.minimumReleaseAge=0 @earendil-works/pi-coding-agent@0.83.0...
[WARN] 1 deprecated subdependencies found: node-domexception@1.0.0
Packages: +128
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Progress: resolved 137, reused 103, downloaded 25, added 28, done

global:
+ @earendil-works/pi-coding-agent 0.83.0

Done in 4.2s using pnpm v11.8.0
Updated pi from 0.82.1 to 0.83.0

pi coding agent的配置

全局配置放在 ~/.pi/agent/settings.json 文件，影响所有项目。
项目配置放在项目目录的 .pi/settings.json 中，只覆盖当前项目。
项目配置中的嵌套对象会与全局配置合并，而非整体替换。

除了两层 settings.json，Pi 还使用了四种专门化的配置文件，各有不同的用途：

配置文件	位置	用途	加载时机
AGENTS.md	项目根目录或 ~/.pi/agent/	项目上下文与编码指令，注入 system prompt	启动时自动加载
SYSTEM.md	~/.pi/agent/	全局行为规则，追加到 system prompt 尾部	启动时加载
settings.json	全局或 .pi/	模型选择、UI 主题、压缩策略、重试等运行参数	启动时加载
models.json	~/.pi/agent/	自定义模型与 Provider（Ollama、vLLM 等）	每次打开 /model 时重新加载
auth.json	~/.pi/agent/	API Key 和 OAuth 凭据（权限 0600）

以下是我让其他AI Agent帮我生成的两个配置文件，基于阿里云dashscope的kimi-k3，仅供参考：

首先配置~/.pi/agent/settings.json，配置参考如下：

{
  "lastChangelogVersion": "0.83.0",
  "theme": "dark",
  "defaultProvider": "dashscope",
  "defaultModel": "kimi/kimi-k3"
}

其次配置~/.pi/agent/models.json，配置参考如下：

  "providers": {
    "dashscope": {
      "baseUrl": "https://XXXXXXXXXX.cn-beijing.maas.aliyuncs.com/compatible-mode/v1",
      "api": "openai-completions",
      "apiKey": "sk-ws-XXXXXXXXXXX",
      "compat": {
        "supportsDeveloperRole": false,
        "supportsReasoningEffort": false,
        "deferredToolsMode": "kimi"
      },
      "models": [
        {
          "id": "kimi/kimi-k3",
          "name": "Kimi K3 (DashScope)",
          "reasoning": true,
          "input": ["text"],
          "contextWindow": 256000,
          "maxTokens": 32768,
          "cost": {
            "input": 0,
            "output": 0,
            "cacheRead": 0,
            "cacheWrite": 0
          }
        }
      ]
    }
  }
}

开始使用

输入pi命令，开始对话

pi

在pi命令中，有一些基础的命令，参考如下：

Command	Description
/login, /logout	Manage OAuth or API-key credentials
/llama	Download, load, and unload llama.cpp router models
/model	Switch models
/scoped-models	Enable/disable models for Ctrl+P cycling
/settings	Thinking level, theme, message delivery, transport
/resume	Pick from previous sessions
/new	Start a new session
/name <name>	Set session display name
/session	Show session file, ID, messages, tokens, and cost
/tree	Jump to any point in the session and continue from there
/trust	Save project trust decision for future sessions
/fork	Create a new session from a previous user message
/clone	Duplicate the current active branch into a new session
/compact [prompt]	Manually compact context, optionally with custom instructions
/copy	Copy last assistant message to clipboard
/export [file]	Export session to HTML or JSONL
/import <file>	Import and resume a session from a JSONL file
/share	Upload as private GitHub gist with shareable HTML link
/reload	Reload keybindings, extensions, skills, prompts, themes, and context files
/hotkeys	Show all keyboard shortcuts
/changelog	Display version history
/quit	Quit pi

参考文档

Pi Documentation · Documentation · Pi
Pi Agent 配置进阶：AGENTS.md、模型切换与 Thinking Level 实战指南 | IT小灶