3 Commits

Include renames

Author	SHA1	Message	Date
slahiriandClaude Opus 4.5	ec05fb561f	fix: detect and clean garbage LLM output ... - Add remove_runaway_repetitions() to clean "dark, dark, dark..." patterns - Add is_garbage_output() to detect LLM generation failures - LLM mode now falls back to NLP output when garbage is detected - Limits any word to max 3 repetitions in output 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>	2025-12-27 20:52:01 +05:30
slahiriandClaude Opus 4.5	d11328ff90	feat: add STRUCTURED prompt style with category-segregated output ... - Add STRUCTURED style that outputs each category on separate lines - Fix '3d' token stripping - only strip booru patterns (1girl, 2boys) - Increase max_tokens_per_category to 50 (was 5) - Increase validator limits to preserve more information - Preserve newlines throughout validation pipeline - Disable truncation and duplicate removal by default 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>	2025-12-27 19:05:47 +05:30
slahiriandClaude Opus 4.5	700607305b	feat(compose): implement Phase 5 - Prompt Validation ... Adds the validator module for quality analysis of assembled prompts: - base.py: ValidatorConfig, ValidationResult, ValidationIssue classes - rules.py: Conflict detection, duplicate finding, quality checks - standard.py: StandardValidator with auto-correction Validates: - Length constraints (CLIP token limits) - Duplicate words/phrases detection - Conflicting terms (e.g., girl vs woman) - Missing required categories - Quality issues (punctuation, spacing) 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>	2025-12-27 11:41:40 +05:30