# CHANGES
- Add `apply_ul_tags` pattern for content categorization
- Add `extract_mcp_servers` pattern for MCP server identification
- Add `generate_code_rules` pattern for AI coding guardrails
- Add `t_check_dunning_kruger` pattern for competence assessment
- Renumber all patterns from 37-226 to 37-230
- Insert new patterns at positions 37, 129, 153, 203
- Upgrade `anthropic-sdk-go` to version 1.16.0
- Remove outdated model `ModelClaude3_5SonnetLatest`
- Add new model `ModelClaudeSonnet4_5_20250929`
- Include `ModelClaudeSonnet4_5_20250929` in `modelBetas` map
### CHANGES
- Add `create_conceptmap` for HTML concept maps using Vis.js
- Introduce `fix_typos` for text proofreading and corrections
- Implement `model_as_sherlock_freud` for psychological modeling
- Add `predict_person_actions` for behavior prediction
- Include `recommend_yoga_practice` for personalized yoga guidance
- Credit pattern contribution to @FELIPEGUEDESBR
### CHANGES
- Add `fix_typos` for proofreading and correcting errors
- Introduce `model_as_sherlock_freud` for psychological modeling
- Implement `predict_person_actions` for behavioral response predictions
- Add `recommend_yoga_practice` for personalized yoga guidance
- Include `fix_typos` method for text correction
- Add `model_as_sherlock_freud` for behavior analysis
- Introduce `predict_person_actions` for action prediction
# CHANGES
- Add new WELLNESS category with four patterns
- Add `model_as_sherlock_freud` for psychological detective analysis
- Add `predict_person_actions` for behavioral response predictions
- Add `recommend_yoga_practice` for personalized wellness guidance
- Add `fix_typos` pattern for proofreading corrections
- Update ANALYSIS category to include new patterns
- Update SELF category with wellness-related patterns
- Tag existing patterns with WELLNESS classification
- Add prominent warning at top of Extensions guide with visual indicators
- Update main README with brief Extensions section and link to full guide
- Remove misleading examples showing direct piping to fabric
- Add clear examples: ❌ what DOES NOT WORK vs ✅ what WORKS
- Consolidate all extension documentation in Examples/README.md
- Explain technical reason: extensions only processed via ApplyTemplate()
- Prevents user confusion about extension syntax processing
- Extract InputSentinel constant to shared constants.go file
- Remove duplicate inputSentinel definitions from template.go and patterns.go
- Create withTestExtension helper function to reduce test code duplication
- Refactor 3 test functions to use the helper (reduces ~40 lines per test)
- Fix shell script to use $@ instead of $* for proper argument quoting
Addresses review comments from @ksylvan and @Copilot AI
### CHANGES
- Add `extract_characters` to identify and describe characters
- Update business category to include `extract_characters`
- Include `extract_characters` in extract category
- Add `extract_characters` description in pattern descriptions JSON
- Update user documentation with `extract_characters` details
## CHANGES
- Add top-level project name to navigation hierarchy
- Nest all sections under main project heading
- Fix npm install script path extension
- Update localhost URL to use HTML format
- Add "Mdsvex" to VSCode spelling dictionary
- Include "details" and "summary" to HTML tags
- Remove trailing newline from web README
- Remove duplicate fabric-summarize.png screenshot
- Wrap Updates section in HTML details/summary accordion to save space
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
CHANGES
- Define character extraction goals and steps
- Specify canonical naming and deduplication rules
- Outline interaction mapping and narrative importance
- Provide output schema with formatting guidelines
- Include positive/negative examples for clarity
- Enforce no speculative motivations or non-actors
- Set fallback for no characters found
