Merge pull request #28 from second-state/skill

Improve the SKILLS

Author: L-jasmine

.DS_Store

@@ -0,0 +1,82 @@
+# Changelog
+
+All notable changes to this project will be documented in this file.
+
+The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
+and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
+
+## [1.3.1] - 2025-01-16
+
+### Changed
+- Added `export RUST_LOG=debug` before launching server in Phase 5, Step 6
+- Server now runs with debug logging enabled for better troubleshooting
+
+## [1.3.0] - 2025-01-16
+
+### Fixed
+- **CRITICAL**: Corrected config.toml section order to `[tts]` → `[asr]` → `[llm]` (required by EchoKit server parser)
+- **CRITICAL**: Fixed platform-specific field names:
+  - ElevenLabs TTS now uses `token` instead of `api_key`
+  - ElevenLabs TTS now uses `model_id` instead of `model`
+- Removed header comments from config.toml (parser rejects comments at file start)
+- Added required ASR fields: `prompt` and `vad_url` for Whisper compatibility
+- Updated platform YAML metadata with `api_key_field` and `model_field` properties
+
+### Changed
+- Enhanced Phase 4 documentation with platform-specific field mapping guide
+- Enhanced Phase 5 with warnings about correct field names when updating API keys
+- Improved error prevention by documenting exact field name requirements per platform
+
+## [1.2.0] - 2025-01-15
+
+### Added
+- Phase 5: API Key Entry and Server Launch
+  - Interactive API key collection for all services
+  - Automatic config.toml update with provided keys
+  - Server build verification and automatic building
+  - Server launch with background process management
+  - Local IP address detection for WebSocket URL display
+  - Process ID tracking for server management
+- Enhanced success messages with actual WebSocket URLs
+- Server verification steps after launch
+- Comprehensive error handling for build failures and server crashes
+
+### Changed
+- Updated workflow from file generation only to complete server setup
+- Five-phase process instead of four-phase
+
+## [1.1.0] - 2025-01-14
+
+### Added
+- Enhanced system prompt generation with tone-based behavioral defaults
+- Custom platform support with auto-configuration via WebSearch
+- Corrected MCP server configuration format
+- Platform auto-detection for common providers (Groq, DeepSeek, etc.)
+
+### Changed
+- Improved system prompt sophistication with domain knowledge integration
+- Better handling of user constraints and formatting requirements
+
+## [1.0.0] - 2025-01-13
+
+### Added
+- Initial release of EchoKit Config Generator SKILL
+- Four-phase interactive configuration flow
+- Platform knowledge base with 9+ platforms:
+  - ASR: OpenAI Whisper, Local Whisper, Deepgram
+  - TTS: OpenAI, ElevenLabs, Azure Speech
+  - LLM: OpenAI GPT, Anthropic Claude, Ollama
+- System prompt generation from user requirements
+- MCP server templates (filesystem, git, custom)
+- Example configurations for common use cases
+- Comprehensive documentation
+- Zero external dependencies (completely standalone)
+
+### Features
+- Interactive platform selection with API key locations
+- Automatic config.toml generation
+- SETUP_GUIDE.md with step-by-step instructions
+- MCP server support (optional)
+- Pre-built examples (voice companion, coding assistant)
+
+[1.0.0]: https://github.com/YOUR_USERNAME/echokit-config-skill/releases/tag/v1.0.0

.claude/.DS_Store

@@ -30,13 +30,6 @@ This SKILL can be installed in the following ways:
 git clone https://github.com/second-state/echokit_server.git
 cd echokit_server
 
-# Create symlink to Claude skills directory
-ln -s $(pwd)/skills/echokit-config-generator ~/.claude/skills/echokit-config-generator
-
-# Or copy directly
-cp -r skills/echokit-config-generator ~/.claude/skills/
-```
-
 **That's it!** The SKILL is now available in Claude Code.
 
 ### Usage