Audio Cog

- Added environment requirements for the skill: now specifies needed binaries (python3) and environment variables (CELLCOG_API_KEY) in metadata. - No user-facing feature or functionality changes; documentation in SKILL.md updated to reflect setup prerequisites.

- Updated documentation for clarity and accuracy in SKILL.md - Improved and expanded description, highlighting support for podcasts and dialogue - Clarified agent usage: now specifies “all agents except OpenClaw” for blocking chat integration example - Refined instructions and language throughout for easier onboarding and provider selection - No code or functional changes—documentation update only

- Improved documentation in SKILL.md with more concise descriptions and clearer usage instructions. - Expanded SDK usage code sample to explicitly show client initialization. - Updated skill description to highlight avatar voices and music generation up to 10 minutes. - Enhanced formatting and clarified steps for using voice providers, avatars, sound effects, and music features.

- Simplified and clarified the description to emphasize major features and use cases. - Reorganized usage instructions for easier onboarding, highlighting SDK references up front. - Added new "If CellCog is not installed" section with agent-specific installation guidance. - Streamlined sections for voice providers and capabilities; removed duplicative wording. - Made instructions for different agent types (OpenClaw, Cursor, etc.) more concise and highlighted code output. - Shortened and focused provider feature explanations for improved readability.

- Expanded SKILL.md with detailed guidance on provider selection (OpenAI, ElevenLabs, MiniMax) and their strengths. - Added new tables outlining voice provider scenarios, voice options, customization tips, and emotion tag usage. - Provided explicit examples for avatar/cloned voices and usage of custom avatars for personalized narration. - Enhanced sections on sound effects and music generation, including sample prompts and best practice tips. - Clarified multi-language support and practical agent usage for audio generation. - Included a new "Tips for Better Audio" section to help users optimize results with the skill.

- Expanded description and usage details for TTS, music, SFX, and podcast production features - Clarified role of supported providers: OpenAI, ElevenLabs, and MiniMax, with highlights for multi-voice and avatar voices - Documented new features: multi-voice dialogue, podcast pipeline, 160+ voices, and output in MP3/WAV - Updated related skills section for audio, music, podcast, and video generation - Refined and reorganized documentation for easier discovery of key capabilities

audio-cog 1.0.6 - Added explicit Python code examples for agent-based and blocking audio generation using the SDK. - Clarified that OpenClaw agent mode is recommended for long tasks and provided notify_session_key details. - Referenced the main cellcog skill for advanced SDK usage, delivery modes, and file handling. - Documentation updates only; no functional or interface changes.

- Added OS compatibility metadata for Darwin, Linux, and Windows. - Updated skill description for improved clarity and SEO. - Added homepage link to CellCog website. - Improved formatting and metadata structure in SKILL.md. - No changes to core functionality; documentation and metadata only.

- Adds support for three voice providers: OpenAI, ElevenLabs, and MiniMax, each with unique capabilities. - Introduces avatar/cloned voice generation via MiniMax, allowing users to create audio in their own voice. - Expands features to include standalone sound effects (up to 30 seconds) and longer music generation (up to 10 minutes). - Clarifies provider recommendations by scenario, with detailed guidance on emotional tags (ElevenLabs) and fine-grained controls (MiniMax). - Updates documentation with new usage examples, usage tips, and multi-language support across all providers.

- Added author and dependencies fields to SKILL.md for clearer metadata. - Updated prerequisite instructions to refer to the cellcog skill by name. - Minor edits for clarity and consistency in setup and usage guidance.

audio-cog 1.0.2 - Added detailed documentation of all 8 available CellCog voices, including usage recommendations and voice characteristics. - Included guidance on choosing voices by content type and how to customize styles (accent, emotion, pacing, etc.). - Added music licensing statement: all generated music is royalty-free and usable for any commercial purpose. - Expanded multi-language support list and updated multi-language example prompts. - Updated example prompts and tips to reflect voice selection and new usage patterns.

- Adds metadata (including an emoji) for enhanced identification. - Updates quick-start usage pattern to use `create_chat` with simplified, fire-and-forget execution and notification model (v1.0+). - Clearly standardizes on `chat_mode="agent"` as optimal for all audio tasks, deprecating previous agent team recommendations. - Updates guidance to reflect new workflow and best practices. - No change in feature set or audio capabilities.

- Initial release of audio-cog: Professional AI audio generation powered by CellCog. - Supports text-to-speech, voice synthesis, narration, voiceovers, podcast production, music creation, and sound design. - Offers voice customization (gender, age, emotion, accent, pacing, tone). - Enables music and background audio generation with detailed control (genre, tempo, mood, instruments, duration). - Multi-language speech generation and various audio output formats. - Includes prompt examples, guidance for agent team mode, and detailed usage tips.