@kikkimo/claude-launcher 3.1.0 → 3.2.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
package/CHANGELOG.md CHANGED
@@ -5,6 +5,21 @@ All notable changes to this project will be documented in this file.
5
5
  The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
6
6
  and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
7
7
 
8
+ ## [3.2.0] - 2026-06-14
9
+
10
+ ### Added
11
+ - **New Flagship Models**: `claude-opus-4-8` (anthropic), `kimi-k2.7-code` (moonshot), `MiniMax-M3` (minimax), `glm-5.2[1m]` (zhipu/zai).
12
+ - **Provider Template Drift Migration**: On first load after a provider template upgrade, saved API configs now auto-refresh tier defaults (`ANTHROPIC_DEFAULT_OPUS_MODEL` etc.) while preserving user manual overrides. Backed by `_normalizeApiFields` drift detection; persistence flows through the existing `_migrateApiEntry` → constructor `saveConfig()` lifecycle.
13
+
14
+ ### Changed
15
+ - **GLM Fixed Tier Template**: zhipu/zai now use a fixed tier template (`Opus = Sonnet = glm-5.2[1m]`, `Haiku = glm-5-turbo`) regardless of selected model, replacing the old `glm-5.1 → glm-5-turbo` fast map. `glm-5.1` retained as an optional model (no upgrade alias).
16
+ - **DeepSeek Haiku Alignment**: Haiku tier standardized to `deepseek-v4-flash` (dropped `[1m]`); legacy `deepseek-v4-flash[1m]` configs auto-upgrade via alias.
17
+ - **Moonshot Single Flagship**: Converged to `kimi-k2.7-code` as the sole model; legacy `kimi-k2.6`/`k2.5`/`k2-thinking`/`k2-thinking-turbo`/preview variants now upgrade aliases. Added `ENABLE_TOOL_SEARCH: 'false'` and `CLAUDE_CODE_AUTO_COMPACT_WINDOW: '262144'` (provider-only defaults; override via Custom Vars).
18
+ - **GLM Context Window**: zhipu/zai added `CLAUDE_CODE_AUTO_COMPACT_WINDOW: '1000000'` to match the 1M context of `glm-5.2[1m]`.
19
+
20
+ ### Removed
21
+ - **Moonshot Main List**: Removed `kimi-k2.5`/`kimi-k2-thinking`/`kimi-k2-thinking-turbo` from the model list (retained as upgrade aliases).
22
+
8
23
  ## [3.1.0] - 2026-05-09
9
24
 
10
25
  ### Added
package/README.md CHANGED
@@ -28,10 +28,10 @@ An elegant interactive launcher for Claude Code with a beautiful Claude-style in
28
28
  - Strong password requirements and validation
29
29
 
30
30
  ### 🚀 **Third-party API Management**
31
- - Full support for 9 third-party API providers (Anthropic, DeepSeek, Kimi K2.6, MiniMax M2.7, GLM-5.1/ZhiPu AI, and custom APIs)
31
+ - Full support for 9 third-party API providers (Anthropic, DeepSeek, Kimi K2.7-Code, MiniMax M3, GLM-5.2/ZhiPu AI, and custom APIs)
32
32
  - **6-Step Add API Wizard**: Provider → URL → Token → Model → Name → Config Confirm with pre-create duplicate detection, back navigation between steps, and inline config editing before persist
33
33
  - **Env Config Editor**: Edit model config (6 fields: Sonnet/Opus/Haiku/Subagent/Custom), runtime config (6 fields: timeout/attribution/nonessential/effort/experimental/nonstreaming), and custom env vars — with per-field hints, provider default values, and overridden markers
34
- - **Auto Model Tier Matching**: Same-generation auto-matching for all providers (AnthropicOpus/Sonnet/Haiku, DeepSeek pro/flash, GLM 5.1/turbo, etc.)
34
+ - **Auto Model Tier Matching**: Same-generation auto-matching for all providers (Anthropic Opus/Sonnet/Haiku, DeepSeek pro/flash, GLM 5.2/turbo fixed tiers, etc.)
35
35
  - **Interactive API Editing**: Modify name, provider, base URL, model, and all env configs for existing APIs
36
36
  - Dynamic column alignment and comprehensive field hints across all supported languages
37
37
  - API usage statistics with success/failure tracking
@@ -183,7 +183,7 @@ Claude Launcher uses an advanced configuration system:
183
183
 
184
184
  Configure any third-party API provider through the interactive interface:
185
185
 
186
- - **Supported Providers**: Anthropic (Opus 4.7/Sonnet 4.6/Haiku 4.5), DeepSeek (V4-Pro/V4-Flash), Moonshot/Kimi (K2.6), MiniMax CN/Global (M2.7), ZhiPu AI/Z.ai (GLM-5.1/5-Turbo), and custom Anthropic-compatible APIs
186
+ - **Supported Providers**: Anthropic (Opus 4.8/Sonnet 4.6/Haiku 4.5), DeepSeek (V4-Pro/V4-Flash), Moonshot/Kimi (K2.7-Code), MiniMax CN/Global (M3), ZhiPu AI/Z.ai (GLM-5.2/5-Turbo), and custom Anthropic-compatible APIs
187
187
  - **Secure Storage**: All API tokens encrypted before storage
188
188
  - **Validation**: Real-time validation of URLs, tokens, and models
189
189
  - **Usage Tracking**: Monitor API usage statistics with success/failure rates
package/docs/README-zh.md CHANGED
@@ -28,10 +28,10 @@
28
28
  - 强密码要求和验证
29
29
 
30
30
  ### 🚀 **第三方 API 管理**
31
- - 全面支持 9 个第三方 API 提供商(Anthropic、DeepSeek、Kimi K2.6、MiniMax M2.7、GLM-5.1/智谱AI 和自定义 API)
31
+ - 全面支持 9 个第三方 API 提供商(Anthropic、DeepSeek、Kimi K2.7-Code、MiniMax M3、GLM-5.2/智谱AI 和自定义 API)
32
32
  - **6 步新增 API 向导**:提供商 → URL → Token → 模型 → 名称 → 配置确认,支持预创建重复检测、步骤间返回导航、保存前内联编辑配置
33
33
  - **环境变量配置编辑**:编辑模型配置(6 项:Sonnet/Opus/Haiku/子代理/自定义)、运行配置(6 项:超时/署名/非必要流量/推理强度/实验性功能/非流式回退)和自定义变量——配有逐字段提示、Provider 推荐默认值和已覆盖标记
34
- - **模型同代自动匹配**:所有 Provider 智能匹配同代档位(Anthropic Opus/Sonnet/Haiku、DeepSeek pro/flash、GLM 5.1/turbo 等)
34
+ - **模型同代自动匹配**:所有 Provider 智能匹配同代档位(Anthropic Opus/Sonnet/Haiku、DeepSeek pro/flash、GLM 5.2/turbo 固定档位 等)
35
35
  - **交互式 API 编辑**:修改已有 API 的名称、供应商、URL、模型及全部环境变量配置
36
36
  - 动态列对齐和全面的字段提示,支持全部 11 种语言
37
37
  - API 使用统计,支持成功/失败率追踪
@@ -183,7 +183,7 @@ Claude Launcher 使用先进的配置系统:
183
183
 
184
184
  通过交互界面配置任何第三方 API 提供商:
185
185
 
186
- - **支持的提供商**:Anthropic(Opus 4.7/Sonnet 4.6/Haiku 4.5)、DeepSeek(V4-Pro/V4-Flash)、Moonshot/Kimi(K2.6)、MiniMax 国内版/国际版(M2.7)、智谱AI/Z.ai(GLM-5.1/5-Turbo)和自定义 Anthropic 兼容 API
186
+ - **支持的提供商**:Anthropic(Opus 4.8/Sonnet 4.6/Haiku 4.5)、DeepSeek(V4-Pro/V4-Flash)、Moonshot/Kimi(K2.7-Code)、MiniMax 国内版/国际版(M3)、智谱AI/Z.ai(GLM-5.2/5-Turbo)和自定义 Anthropic 兼容 API
187
187
  - **安全存储**:所有 API 令牌在存储前加密
188
188
  - **验证**:URL、令牌和模型的实时验证
189
189
  - **使用跟踪**:监控 API 使用统计,支持成功/失败率追踪