harness-auto-docs 0.3.1 → 0.3.3

package/AGENTS.md

@@ -63,7 +63,15 @@ doc targets, calls an LLM (Claude or GPT), writes Markdown files, and opens a Gi
 | Variable | Description |
 |----------|-------------|
 | `AI_MODEL` | Model name: `claude-sonnet-4-6`, `gpt-4o`, etc. |
-| `AI_API_KEY` | Anthropic or OpenAI API key |
+| `ANTHROPIC_API_KEY` | Anthropic API key (required for `claude-*` models) |
+| `OPENAI_API_KEY` | OpenAI API key (required for `gpt-*` models) |
+| `MINIMAX_API_KEY` | MiniMax API key (required for `MiniMax-*` models) |
+| `QWEN_API_KEY` | Alibaba Qwen API key (required for `qwen-*` models) |
+| `ZHIPU_API_KEY` | Zhipu AI key (required for `glm-*` models) |
+| `DEEPSEEK_API_KEY` | DeepSeek API key (required for `deepseek-*` models) |
+| `DOUBAO_API_KEY` | Doubao API key (required for `doubao-*` models) |
+| `KIMI_API_KEY` | Moonshot AI key (required for `moonshot-*` models) |
+| `GROK_API_KEY` | xAI Grok key (required for `grok-*` models) |
 | `GITHUB_TOKEN` | GitHub PAT or Actions token |
 
 ## How to extend

package/README.md

@@ -1,3 +1,6 @@
+[![translation](https://img.shields.io/badge/translation->中文-blue?style=flat-square&logo=googletranslate)](README_zh.md)
+[![translation](https://img.shields.io/badge/translation->日本語-blue?style=flat-square&logo=googletranslate)](README_ja.md)
+
 # harness-auto-docs
 
 Auto-generate [Harness Engineering](https://openai.com/research/harness-engineering) style documentation when your project creates a git tag. Opens a PR with updated docs.
@@ -6,7 +9,7 @@ Auto-generate [Harness Engineering](https://openai.com/research/harness-engineer
 
 Add `.github/workflows/harness-docs.yml` to your project (see `examples/github-workflow.yml`), then set:
 
-- `AI_API_KEY` — your Anthropic, OpenAI, or MiniMax API key (repository secret)
+- `ANTHROPIC_API_KEY` / `OPENAI_API_KEY` / `MINIMAX_API_KEY` / … — API key for the chosen provider (see Supported models below)
 - `AI_MODEL` — model name, e.g. `claude-sonnet-4-6`, `gpt-4o`, or `MiniMax-Text-01`
 - `GITHUB_TOKEN` — provided automatically by GitHub Actions
 
@@ -34,18 +37,24 @@ When you push a tag (`git tag v1.2.0 && git push --tags`), a PR is automatically
 ## Local run
 
 ```bash
-AI_MODEL=claude-sonnet-4-6 AI_API_KEY=sk-ant-... GITHUB_TOKEN=ghp_... npx harness-auto-docs
+AI_MODEL=claude-sonnet-4-6 ANTHROPIC_API_KEY=sk-ant-... GITHUB_TOKEN=ghp_... npx harness-auto-docs
 ```
 
 ## Supported models
 
 Model routing is determined by the prefix of `AI_MODEL`:
 
-| Prefix | Provider | Example models |
-|--------|----------|----------------|
-| `claude-*` | Anthropic | `claude-sonnet-4-6`, `claude-opus-4-6`, `claude-haiku-4-5-20251001` |
-| `gpt-*` | OpenAI | `gpt-4o`, `gpt-4o-mini`, `o3` |
-| `MiniMax-*` | MiniMax (Anthropic-compatible) | `MiniMax-Text-01` |
+| Prefix | Provider | API key env var | Example models |
+|--------|----------|-----------------|----------------|
+| `claude-*` | Anthropic | `ANTHROPIC_API_KEY` | `claude-sonnet-4-6`, `claude-opus-4-6` |
+| `gpt-*` | OpenAI | `OPENAI_API_KEY` | `gpt-4o`, `gpt-4o-mini`, `o3` |
+| `MiniMax-*` | MiniMax | `MINIMAX_API_KEY` | `MiniMax-Text-01` |
+| `qwen-*` | Qwen (阿里云) | `QWEN_API_KEY` | `qwen-turbo`, `qwen-plus`, `qwen-max` |
+| `glm-*` | 智谱 | `ZHIPU_API_KEY` | `glm-4`, `glm-4-flash` |
+| `deepseek-*` | DeepSeek | `DEEPSEEK_API_KEY` | `deepseek-chat`, `deepseek-coder` |
+| `doubao-*` | 豆包 (ByteDance) | `DOUBAO_API_KEY` | `doubao-pro-4k` |
+| `moonshot-*` | Kimi (Moonshot) | `KIMI_API_KEY` | `moonshot-v1-8k`, `moonshot-v1-32k` |
+| `grok-*` | Grok (xAI) | `GROK_API_KEY` | `grok-2`, `grok-beta` |
 
 ## Future
 

package/README_ja.md

@@ -0,0 +1,61 @@
+[![translation](https://img.shields.io/badge/translation->English-blue?style=flat-square&logo=googletranslate)](README.md)
+[![translation](https://img.shields.io/badge/translation->中文-blue?style=flat-square&logo=googletranslate)](README_zh.md)
+
+# harness-auto-docs
+
+プロジェクトの git タグ作成時に、[Harness Engineering](https://openai.com/research/harness-engineering) スタイルのドキュメントを自動生成し、PR を開いてドキュメントを更新します。
+
+## 使用方法
+
+`.github/workflows/harness-docs.yml` をプロジェクトに追加し（`examples/github-workflow.yml` を参照）、以下を設定します：
+
+- `ANTHROPIC_API_KEY` / `OPENAI_API_KEY` / `MINIMAX_API_KEY` / … — 選択したプロバイダーの API キー（詳細はサポートモデル表を参照）
+- `AI_MODEL` — モデル名（例：`claude-sonnet-4-6`、`gpt-4o`、`MiniMax-Text-01`）
+- `GITHUB_TOKEN` — GitHub Actions により自動提供
+
+タグをプッシュすると（`git tag v1.2.0 && git push --tags`）、自動的にドキュメントを更新する PR が開きます。
+
+## 更新されるファイル
+
+| ファイル | 常時更新 | 条件付き更新 |
+|---------|---------|-------------|
+| `AGENTS.md` | ✓ | |
+| `ARCHITECTURE.md` | ✓ | |
+| `docs/DESIGN.md` | ✓ | |
+| `docs/QUALITY_SCORE.md` | ✓ | |
+| `changelog/vX.Y.Z.md` | ✓ | |
+| `docs/design-docs/vX.Y.Z.md` | ✓ | |
+| `docs/design-docs/index.md` | ✓ | |
+| `docs/exec-plans/tech-debt-tracker.md` | ✓ | |
+| `docs/FRONTEND.md` | | フロントエンドファイル変更時 |
+| `docs/SECURITY.md` | | 認証/セキュリティファイル変更時 |
+| `docs/RELIABILITY.md` | | インフラファイル変更時 |
+| `docs/generated/db-schema.md` | | SQL/schema ファイル変更時 |
+| `docs/product-specs/index.md` | | 新機能検出時 |
+| `docs/references/` | | 新規依存関係追加時 |
+
+## ローカル実行
+
+```bash
+AI_MODEL=claude-sonnet-4-6 ANTHROPIC_API_KEY=sk-ant-... GITHUB_TOKEN=ghp_... npx harness-auto-docs
+```
+
+## サポートモデル
+
+モデルルーティングは `AI_MODEL` のプレフィックスで決定されます：
+
+| プレフィックス | プロバイダー | API キー環境変数 | モデル例 |
+|--------------|------------|----------------|---------|
+| `claude-*` | Anthropic | `ANTHROPIC_API_KEY` | `claude-sonnet-4-6`、`claude-opus-4-6` |
+| `gpt-*` | OpenAI | `OPENAI_API_KEY` | `gpt-4o`、`gpt-4o-mini`、`o3` |
+| `MiniMax-*` | MiniMax | `MINIMAX_API_KEY` | `MiniMax-Text-01` |
+| `qwen-*` | Qwen（阿里云） | `QWEN_API_KEY` | `qwen-turbo`、`qwen-plus`、`qwen-max` |
+| `glm-*` | 智谱 AI | `ZHIPU_API_KEY` | `glm-4`、`glm-4-flash` |
+| `deepseek-*` | DeepSeek | `DEEPSEEK_API_KEY` | `deepseek-chat`、`deepseek-coder` |
+| `doubao-*` | 豆包（ByteDance） | `DOUBAO_API_KEY` | `doubao-pro-4k` |
+| `moonshot-*` | Kimi（Moonshot） | `KIMI_API_KEY` | `moonshot-v1-8k`、`moonshot-v1-32k` |
+| `grok-*` | Grok（xAI） | `GROK_API_KEY` | `grok-2`、`grok-beta` |
+
+## 今後の予定
+
+- GitLab サポート（MR 作成）

package/README_zh.md

@@ -0,0 +1,61 @@
+[![translation](https://img.shields.io/badge/translation->English-blue?style=flat-square&logo=googletranslate)](README.md)
+[![translation](https://img.shields.io/badge/translation->日本語-blue?style=flat-square&logo=googletranslate)](README_ja.md)
+
+# harness-auto-docs
+
+当项目创建 git tag 时，自动生成 [Harness Engineering](https://openai.com/research/harness-engineering) 风格的文档，并自动开启 PR 更新文档。
+
+## 使用方法
+
+将 `.github/workflows/harness-docs.yml` 添加到项目中（参见 `examples/github-workflow.yml`），然后设置：
+
+- `ANTHROPIC_API_KEY` / `OPENAI_API_KEY` / `MINIMAX_API_KEY` / … — 所选供应商的 API 密钥（详见下方支持的模型表格）
+- `AI_MODEL` — 模型名称，如 `claude-sonnet-4-6`、`gpt-4o` 或 `MiniMax-Text-01`
+- `GITHUB_TOKEN` — GitHub Actions 自动提供
+
+当你推送标签（`git tag v1.2.0 && git push --tags`）时，系统会自动开启一个包含更新文档的 PR。
+
+## 更新范围
+
+| 文件 | 始终更新 | 条件更新 |
+|------|---------|---------|
+| `AGENTS.md` | ✓ | |
+| `ARCHITECTURE.md` | ✓ | |
+| `docs/DESIGN.md` | ✓ | |
+| `docs/QUALITY_SCORE.md` | ✓ | |
+| `changelog/vX.Y.Z.md` | ✓ | |
+| `docs/design-docs/vX.Y.Z.md` | ✓ | |
+| `docs/design-docs/index.md` | ✓ | |
+| `docs/exec-plans/tech-debt-tracker.md` | ✓ | |
+| `docs/FRONTEND.md` | | 前端文件变更时 |
+| `docs/SECURITY.md` | | 认证/安全文件变更时 |
+| `docs/RELIABILITY.md` | | 基础设施文件变更时 |
+| `docs/generated/db-schema.md` | | SQL/schema 文件变更时 |
+| `docs/product-specs/index.md` | | 检测到新功能时 |
+| `docs/references/` | | 新增依赖时 |
+
+## 本地运行
+
+```bash
+AI_MODEL=claude-sonnet-4-6 ANTHROPIC_API_KEY=sk-ant-... GITHUB_TOKEN=ghp_... npx harness-auto-docs
+```
+
+## 支持的模型
+
+模型路由由 `AI_MODEL` 的前缀决定：
+
+| 前缀 | 供应商 | API 密钥环境变量 | 示例模型 |
+|------|--------|-----------------|---------|
+| `claude-*` | Anthropic | `ANTHROPIC_API_KEY` | `claude-sonnet-4-6`、`claude-opus-4-6` |
+| `gpt-*` | OpenAI | `OPENAI_API_KEY` | `gpt-4o`、`gpt-4o-mini`、`o3` |
+| `MiniMax-*` | MiniMax | `MINIMAX_API_KEY` | `MiniMax-Text-01` |
+| `qwen-*` | 通义千问（阿里云） | `QWEN_API_KEY` | `qwen-turbo`、`qwen-plus`、`qwen-max` |
+| `glm-*` | 智谱 AI | `ZHIPU_API_KEY` | `glm-4`、`glm-4-flash` |
+| `deepseek-*` | DeepSeek | `DEEPSEEK_API_KEY` | `deepseek-chat`、`deepseek-coder` |
+| `doubao-*` | 豆包（字节跳动） | `DOUBAO_API_KEY` | `doubao-pro-4k` |
+| `moonshot-*` | Kimi（月之暗面） | `KIMI_API_KEY` | `moonshot-v1-8k`、`moonshot-v1-32k` |
+| `grok-*` | Grok（xAI） | `GROK_API_KEY` | `grok-2`、`grok-beta` |
+
+## 未来计划
+
+- GitLab 支持（MR 创建）

package/dist/ai/deepseek.d.ts

@@ -0,0 +1,7 @@
+import type { AIProvider } from './interface.js';
+export declare class DeepSeekProvider implements AIProvider {
+    private client;
+    private model;
+    constructor(apiKey: string, model: string);
+    generate(prompt: string): Promise<string>;
+}

package/dist/ai/deepseek.js

@@ -0,0 +1,16 @@
+import OpenAI from 'openai';
+export class DeepSeekProvider {
+    client;
+    model;
+    constructor(apiKey, model) {
+        this.client = new OpenAI({ apiKey, baseURL: 'https://api.deepseek.com' });
+        this.model = model;
+    }
+    async generate(prompt) {
+        const completion = await this.client.chat.completions.create({
+            model: this.model,
+            messages: [{ role: 'user', content: prompt }],
+        });
+        return completion.choices[0].message.content ?? '';
+    }
+}

package/dist/ai/doubao.d.ts

@@ -0,0 +1,7 @@
+import type { AIProvider } from './interface.js';
+export declare class DoubaoProvider implements AIProvider {
+    private client;
+    private model;
+    constructor(apiKey: string, model: string);
+    generate(prompt: string): Promise<string>;
+}

package/dist/ai/doubao.js

@@ -0,0 +1,16 @@
+import OpenAI from 'openai';
+export class DoubaoProvider {
+    client;
+    model;
+    constructor(apiKey, model) {
+        this.client = new OpenAI({ apiKey, baseURL: 'https://ark.cn-beijing.volces.com/api/v3' });
+        this.model = model;
+    }
+    async generate(prompt) {
+        const completion = await this.client.chat.completions.create({
+            model: this.model,
+            messages: [{ role: 'user', content: prompt }],
+        });
+        return completion.choices[0].message.content ?? '';
+    }
+}

package/dist/ai/grok.d.ts

@@ -0,0 +1,7 @@
+import type { AIProvider } from './interface.js';
+export declare class GrokProvider implements AIProvider {
+    private client;
+    private model;
+    constructor(apiKey: string, model: string);
+    generate(prompt: string): Promise<string>;
+}

package/dist/ai/grok.js

@@ -0,0 +1,16 @@
+import OpenAI from 'openai';
+export class GrokProvider {
+    client;
+    model;
+    constructor(apiKey, model) {
+        this.client = new OpenAI({ apiKey, baseURL: 'https://api.x.ai/v1' });
+        this.model = model;
+    }
+    async generate(prompt) {
+        const completion = await this.client.chat.completions.create({
+            model: this.model,
+            messages: [{ role: 'user', content: prompt }],
+        });
+        return completion.choices[0].message.content ?? '';
+    }
+}

package/dist/ai/kimi.d.ts

@@ -0,0 +1,7 @@
+import type { AIProvider } from './interface.js';
+export declare class KimiProvider implements AIProvider {
+    private client;
+    private model;
+    constructor(apiKey: string, model: string);
+    generate(prompt: string): Promise<string>;
+}

package/dist/ai/kimi.js

@@ -0,0 +1,16 @@
+import OpenAI from 'openai';
+export class KimiProvider {
+    client;
+    model;
+    constructor(apiKey, model) {
+        this.client = new OpenAI({ apiKey, baseURL: 'https://api.moonshot.cn/v1' });
+        this.model = model;
+    }
+    async generate(prompt) {
+        const completion = await this.client.chat.completions.create({
+            model: this.model,
+            messages: [{ role: 'user', content: prompt }],
+        });
+        return completion.choices[0].message.content ?? '';
+    }
+}

package/dist/ai/qwen.d.ts

@@ -0,0 +1,7 @@
+import type { AIProvider } from './interface.js';
+export declare class QwenProvider implements AIProvider {
+    private client;
+    private model;
+    constructor(apiKey: string, model: string);
+    generate(prompt: string): Promise<string>;
+}

package/dist/ai/qwen.js

@@ -0,0 +1,16 @@
+import OpenAI from 'openai';
+export class QwenProvider {
+    client;
+    model;
+    constructor(apiKey, model) {
+        this.client = new OpenAI({ apiKey, baseURL: 'https://dashscope.aliyuncs.com/compatible-mode/v1' });
+        this.model = model;
+    }
+    async generate(prompt) {
+        const completion = await this.client.chat.completions.create({
+            model: this.model,
+            messages: [{ role: 'user', content: prompt }],
+        });
+        return completion.choices[0].message.content ?? '';
+    }
+}

package/dist/ai/zhipu.d.ts

@@ -0,0 +1,7 @@
+import type { AIProvider } from './interface.js';
+export declare class ZhipuProvider implements AIProvider {
+    private client;
+    private model;
+    constructor(apiKey: string, model: string);
+    generate(prompt: string): Promise<string>;
+}

package/dist/ai/zhipu.js

@@ -0,0 +1,16 @@
+import OpenAI from 'openai';
+export class ZhipuProvider {
+    client;
+    model;
+    constructor(apiKey, model) {
+        this.client = new OpenAI({ apiKey, baseURL: 'https://open.bigmodel.cn/api/paas/v4' });
+        this.model = model;
+    }
+    async generate(prompt) {
+        const completion = await this.client.chat.completions.create({
+            model: this.model,
+            messages: [{ role: 'user', content: prompt }],
+        });
+        return completion.choices[0].message.content ?? '';
+    }
+}

package/dist/cli.js

@@ -9,6 +9,12 @@ import { GitLabProvider } from './providers/gitlab.js';
 import { AnthropicProvider } from './ai/anthropic.js';
 import { OpenAIProvider } from './ai/openai.js';
 import { MiniMaxProvider } from './ai/minimax.js';
+import { QwenProvider } from './ai/qwen.js';
+import { ZhipuProvider } from './ai/zhipu.js';
+import { DeepSeekProvider } from './ai/deepseek.js';
+import { DoubaoProvider } from './ai/doubao.js';
+import { KimiProvider } from './ai/kimi.js';
+import { GrokProvider } from './ai/grok.js';
 function requireEnv(name) {
     const value = process.env[name];
     if (!value) {
@@ -31,14 +37,26 @@ function detectPlatform() {
     }
     return new GitHubProvider(token);
 }
-function selectAI(model, apiKey) {
+function selectAI(model) {
     if (model.startsWith('claude-'))
-        return new AnthropicProvider(apiKey, model);
+        return new AnthropicProvider(requireEnv('ANTHROPIC_API_KEY'), model);
     if (model.startsWith('gpt-'))
-        return new OpenAIProvider(apiKey, model);
+        return new OpenAIProvider(requireEnv('OPENAI_API_KEY'), model);
     if (model.startsWith('MiniMax-'))
-        return new MiniMaxProvider(apiKey, model);
-    console.error(`Error: Unknown model "${model}". Use a claude-*, gpt-*, or MiniMax-* model.`);
+        return new MiniMaxProvider(requireEnv('MINIMAX_API_KEY'), model);
+    if (model.startsWith('qwen-'))
+        return new QwenProvider(requireEnv('QWEN_API_KEY'), model);
+    if (model.startsWith('glm-'))
+        return new ZhipuProvider(requireEnv('ZHIPU_API_KEY'), model);
+    if (model.startsWith('deepseek-'))
+        return new DeepSeekProvider(requireEnv('DEEPSEEK_API_KEY'), model);
+    if (model.startsWith('doubao-'))
+        return new DoubaoProvider(requireEnv('DOUBAO_API_KEY'), model);
+    if (model.startsWith('moonshot-'))
+        return new KimiProvider(requireEnv('KIMI_API_KEY'), model); // Kimi is the product; Moonshot AI is the provider
+    if (model.startsWith('grok-'))
+        return new GrokProvider(requireEnv('GROK_API_KEY'), model);
+    console.error(`Error: Unknown model "${model}". Supported prefixes: claude-*, gpt-*, MiniMax-*, qwen-*, glm-*, deepseek-*, doubao-*, moonshot-*, grok-*`);
     process.exit(1);
 }
 function getDefaultBranch() {
@@ -52,14 +70,13 @@ function getDefaultBranch() {
 }
 async function main() {
     const model = requireEnv('AI_MODEL');
-    const apiKey = requireEnv('AI_API_KEY');
     const diff = extractDiff();
     if (!diff.raw.trim()) {
         console.log('No diff between tags. Nothing to document.');
         process.exit(0);
     }
     console.log(`Generating docs: ${diff.prevTag} → ${diff.currentTag}`);
-    const ai = selectAI(model, apiKey);
+    const ai = selectAI(model);
     const targets = selectTargets(diff.fileGroups, diff.changedFiles);
     console.log(`Updating ${targets.length} documents: ${targets.join(', ')}`);
     const results = await generateDocs(ai, diff, targets);