Every step,
from the first commit.

- Introduced a comprehensive UI/UX design intelligence skill with 50 styles, 21 palettes, and 50 font pairings. - Added detailed guidelines for accessibility, performance, and interaction design. - Included new datasets for charts, colors, icons, landing page patterns, products, and React performance optimizations. - Enhanced the skill with actionable insights for various project types, ensuring a robust design framework for web and mobile applications.

- Introduced FastAPI templates for creating production-ready applications with async patterns and dependency injection. - Added LLM evaluation strategies for comprehensive performance measurement of LLM applications, including automated metrics and human feedback. - Created a guide for developing high-quality MCP servers, detailing best practices, implementation patterns, and evaluation guidelines. - Included reference files for evaluation and best practices to support MCP server development.

- Introduced a new configuration file (.codex/config.toml) to define model parameters and features for the Codex system. - Configured model type, reasoning effort, output token limit, and auto compact token limit. - Enabled various features including ghost commit, unified execution, and web search requests.

Fix Essentia LoudnessEBUR128 and Flux calls

Guard danceability on short clips

- Removed the Tonality algorithm from the analyzer, simplifying the tonal analysis process. - Updated atonal detection criteria to focus on spectral features, adjusting thresholds for flatness and entropy. - Revised tonality strength calculation to improve accuracy in distinguishing tonal and atonal content. These changes enhance the efficiency of the audio analysis while maintaining robust detection capabilities.

- Updated TonalFeatures model to include optional fields for is_tonal and tonality_strength. - Modified analyze_tonal method to detect atonal content using spectral flatness and entropy. - Improved key detection logic to return null for atonal content and adjusted confidence scaling accordingly. - Updated related interfaces and functions to handle new tonal feature attributes. These changes aim to improve the accuracy of tonal analysis and provide better handling of atonal audio content.

- Updated configuration to set ENABLE_LLM_ANALYSIS to true. - Introduced a new function to extract descriptive tags from filenames, improving differentiation of audio samples. - Enhanced the tagging logic with more granular thresholds for brightness, energy, attack, texture complexity, and harmonic content, allowing for better categorization of audio features. This update aims to improve the accuracy and uniqueness of AI-generated tags.

Issues fixed: 1. Essentia loop detection was too aggressive, incorrectly tagging one-shot samples (like bass hits) as loops 2. AI batch tagging was generating identical tags for different samples Changes to Essentia analyzer (analyzer.py): - Increased minimum duration for loop detection from 0.5s to 1.0s - Added energy consistency check (one-shots decay, loops sustain) - Increased loop confidence thresholds (0.6 -> 0.75) - Added transient count check in classification (one-shots have 1-2 transients) - Classification now requires high loop confidence + multiple transients + sustained energy to classify as 'loop' Changes to AI tagging (ai-tagging.ts): - Lowered duplicate tag threshold from 70% to 30% (triggers individual tagging fallback sooner for better uniqueness) - Improved audio signature with more distinctive features - Made unique traits description more granular (captures subtle differences) - Enhanced batch prompt with clearer instructions about filename hints and sample differentiation - Restructured sample descriptions to prominently show filename keywords and sample type (one-shot vs loop) Co-authored-by: armin.naimi <[email redacted]>

Add migration for extended Essentia fields

Co-authored-by: armin.naimi <[email redacted]>

Cursor/ai tagging differentiation 5b27

Co-authored-by: armin.naimi <[email redacted]>

Co-authored-by: armin.naimi <[email redacted]>

Co-authored-by: armin.naimi <[email redacted]>

Co-authored-by: armin.naimi <[email redacted]>

- Implemented a fallback mechanism for audio analysis that downloads files locally if S3 analysis fails, ensuring robustness in processing. - Added detailed logging for S3 analysis errors, including the response status and error messages for better debugging. - Introduced temporary file management to clean up local downloads after processing, improving resource management. These changes optimize the audio analysis process, providing a seamless experience even in cases of S3 failures.

- Added functionality to analyze audio files directly from S3/R2, eliminating the need for local downloads. - Introduced new methods in the Essentia client for S3 analysis and updated the hybrid analyzer to utilize these methods. - Enhanced the worker to process audio analysis requests via S3, improving efficiency for distributed deployments. - Updated configuration to support S3/R2 settings, ensuring seamless integration with the audio analysis service. These changes optimize the audio analysis workflow, reducing latency and resource usage.

- Added detailed logging for service availability checks, including service URL and enabled status. - Enhanced error handling to log specific error codes and messages for better diagnostics. - Updated conditions to break on additional error scenarios, improving reliability in detecting service unavailability. These changes aim to provide clearer insights into service status and facilitate troubleshooting.

- Modified getServiceUrl function to prepend "http://" if the provided URL lacks a protocol. - Updated config schema to allow ESSENTIA_SERVICE_URL to accept both full URLs and host:port formats. This change enhances flexibility in service URL configuration, particularly for environments like Railway.

Co-authored-by: armin.naimi <[email redacted]>