@khanglvm/llm-router 1.1.1 → 1.3.0

package/CHANGELOG.md

@@ -5,6 +5,46 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [1.3.0] - 2026-03-05
+
+### Added
+- Added Claude Code OAuth subscription provider support end-to-end:
+  - new subscription type: `claude-code`
+  - Claude OAuth constants and runtime request config (`anthropic-beta`, OAuth token endpoint, Claude messages endpoint)
+  - default Claude subscription model seed list for new subscription providers
+- Added CLI support for Claude subscription auth operations:
+  - `llm-router subscription login --subscription-type=claude-code`
+  - `llm-router subscription logout --subscription-type=claude-code`
+  - `llm-router subscription status --subscription-type=claude-code`
+- Added runtime and CLI test coverage for Claude subscription request translation/headers and setup flows.
+
+### Changed
+- Updated subscription probe and provider upsert flow to build type-specific probe payloads:
+  - ChatGPT Codex keeps Responses/Codex probe shape
+  - Claude Code uses Claude messages probe shape
+- Updated subscription config normalization/workflows so default format and model seed list are selected by `subscriptionType`.
+- Updated README and CLI help text/examples to document both supported OAuth subscription types (`chatgpt-codex`, `claude-code`).
+
+### Fixed
+- Fixed new Claude subscription provider creation default model seeding to correctly use Claude defaults instead of ChatGPT defaults.
+
+## [1.2.0] - 2026-03-04
+
+### Added
+- Added Codex Responses API compatibility layer:
+  - request transformation into Codex Responses payload shape
+  - response transformation from Codex responses/SSE events to OpenAI Chat Completions-compatible output
+  - dedicated runtime tests for request + response transformation coverage
+- Added explicit project-level ignore for local `AGENTS.md`.
+
+### Changed
+- Improved TUI/CLI operation reports to user-friendly structured layouts and tables across provider/model-alias/rate-limit/config flows and operational actions.
+- Improved startup/deploy/worker-key/status outputs to avoid raw config variable style and show friendly fields.
+- Updated subscription auth/provider flow behavior and tests for more robust OAuth/Codex subscription handling.
+
+### Fixed
+- Fixed migration/reporting test expectations and summary rendering stability after report format refactor.
+
 ## [1.1.1] - 2026-03-04
 
 ### Fixed

package/README.md

@@ -77,31 +77,43 @@ Then follow this order.
 Flow:
 1. `Config manager`
 2. `Add/Edit provider`
-3. Select provider type:
-   - `standard` -> endpoint + API key + model list
-   - `subscription` -> OAuth profile for predefined ChatGPT Codex models
-4. Enter provider details
+3. Select provider auth mode:
+   - `API Key` -> endpoint + API key + model list
+   - `OAuth` -> browser OAuth + editable model list
+4. For `OAuth`:
+   - Choose subscription provider (`ChatGPT` or `Claude Code`)
+   - Enter Friendly Name and Provider ID
+   - Complete browser OAuth login inside this same flow
+   - Edit model list (pre-filled defaults; you can add/remove)
+   - llm-router live-tests every selected model before save
 5. Save
 
-### 1b) Add Subscription Provider (ChatGPT Codex)
-Commandline example:
+### 1b) Add Subscription Provider (OAuth)
+Commandline examples:
 
 ```bash
+# ChatGPT Codex subscription
 llm-router config \
   --operation=upsert-provider \
   --provider-id=chatgpt \
-  --name="ChatGPT Subscription" \
-  --type=subscription \
-  --subscription-type=chatgpt-codex \
-  --subscription-profile=default
+  --name="GPT Sub" \
+  --type=subscription
 
-llm-router subscription login --profile=default
-llm-router subscription status --profile=default
+# Claude Code subscription
+llm-router config \
+  --operation=upsert-provider \
+  --provider-id=claude-sub \
+  --name="Claude Sub" \
+  --type=subscription \
+  --subscription-type=claude-code
 ```
 
 Notes:
-- `chatgpt-codex` subscription providers use predefined model IDs managed by llm-router releases.
-- No provider API key or endpoint probing is required for this provider type.
+- OAuth login is run during provider upsert (browser flow by default).
+- Supported `subscription-type`: `chatgpt-codex` and `claude-code` (defaults to `chatgpt-codex`).
+- Default model lists are prefilled by subscription type, then editable.
+- Device-code login is available for `chatgpt-codex` only.
+- No provider API key or endpoint probe input is required for subscription mode.
 
 ### 2) Configure Model Fallback (Optional)
 Flow:
@@ -161,6 +173,7 @@ Local endpoints:
 - Unified: `http://127.0.0.1:<port>/route`
 - Anthropic-style: `http://127.0.0.1:<port>/anthropic`
 - OpenAI-style: `http://127.0.0.1:<port>/openai`
+- OpenAI Responses-style: `http://127.0.0.1:<port>/openai/v1/responses` (Codex CLI-compatible)
 
 ## Connect your coding tool
 
@@ -190,6 +203,8 @@ When local server is running:
 
 No stop/start cycle needed.
 
+Config/status outputs are shown in structured table layouts for easier operator review.
+
 ## Cloudflare Worker (Hosted)
 
 Use when you want a hosted endpoint instead of local server.

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@khanglvm/llm-router",
-  "version": "1.1.1",
+  "version": "1.3.0",
   "description": "Single gateway endpoint for multi-provider LLMs with unified OpenAI+Anthropic format and seamless fallback",
   "keywords": [
     "llm-router",