Removed "attempts > 0" check in research/guarded_ai.py that prevented
hints from showing on the first attempt. Matches the fix made to
activity.py for consistent behavior across web app and CLI simulator.
SPEC.yaml fixes:
- Comment out orphaned example code blocks that broke YAML parsing
- Convert progressive hints and dynamic question examples to comments
- Add placeholder keys to maintain valid YAML structure
- All examples now documented but non-executable (reference only)
- Validates with 0 errors
guarded_ai.py refactor (CLI simulator now uses v2.0 features):
- Import activity_utils.py for consistency with activity.py
- Use check_conditions() for advanced metadata conditions (gte, lt, contains, etc.)
- Use filter_content_blocks() for template rendering and conditional blocks
- Use render_template() for dynamic question text with {{variables}}
- Use resolve_conditional_navigation() for if/elif/else navigation
- Use select_weighted_random() for weighted random selection
- Use get_progressive_hint() for progressive hints system
- Create template contexts with built-in variables (current_attempt, etc.)
Benefits:
- Single source of truth for v2.0 logic (activity_utils.py)
- CLI simulator now tests all v2.0 features
- Maintainability: changes to features only need updates in one place
- Consistency: web app and CLI behave identically
All changes validated and tested.
Add comprehensive v2.0 features to enhance activity creation:
Features Implemented:
- Template variables: {{metadata.key}}, {{current_attempt}}, etc.
- Conditional content blocks: show_if conditions for dynamic content
- Advanced metadata conditions: gte, lt, contains, regex, exists operators
- Conditional navigation: if/elif/else branching based on metadata
- Progressive hints system: Auto-display hints based on attempt number
- Weighted random selection: Probabilistic outcomes with custom weights
- Dynamic question text: Questions with template variables
- Built-in attempt counters: Access to current_attempt, max_attempts, attempts_remaining
Files Modified:
- activity.py: Integrated all v2.0 features into activity execution
- activity_utils.py: New utility module for templates and conditions
- activity_yaml_validator.py: Updated validator for v2.0 schema
- CLAUDE.md: Added session persistence and Twitch Plays model docs
- research/SPEC.yaml: Comprehensive v2.0 feature documentation
Added:
- research/activity-test-v2-features.yaml: Test activity demonstrating all features
All changes validated and tested. Zero errors in validator.
- Add random bucket names (emergency, surprise, bonus) to main buckets list
- Fix invalid transition targets to use existing steps
- Add tokens_for_ai to all feedback_prompts (required field)
- Add bonus transition definition
All validation errors resolved - SPEC.yaml now passes validation
Random Bucket System:
- Probabilistic events that trigger alongside user responses
- Random rolls before categorization to prevent AI bias
- Multiple random events can trigger simultaneously
- User bucket processed first, random events layer on top
- Metadata accumulates across all transitions
- Last transition's navigation wins
Implementation:
- activity.py: Core random bucket rolling logic
- activity_yaml_validator.py: Validation for random_buckets config
- research/guarded_ai.py: CLI simulator with random event display
- tests/unit/test_random_buckets.py: 22 comprehensive tests (all passing)
Fashion Empire Enhancement:
- activity40-fashion-empire-backrooms.yaml: Added random events to 4 zones
- fashion_emergency (5%): Urgent crises testing leadership
- creative_opportunity (10%): Breakthroughs rewarding innovation
- surprise_client (5%): VIP visitors recognizing reputation
- Random events enhance gameplay without hijacking user intent
Documentation:
- research/SPEC.yaml: Complete YAML specification with verbose comments
- All metadata operations (string concat, numeric ops, random)
- Random buckets with flow explanation
- Feedback prompts (multi-agent system)
- Processing scripts (pre_script, processing_script)
- Model overrides (classifier_model, feedback_model)
- Termination patterns and best practices
- Validation rules and examples
New Activities:
- activity-nuclear-power-plant-ai.yaml: Nuclear reactor control simulation
- activity-submarine-simulation.yaml: Deep sea exploration
- activity-unwaste-factory.yaml: Recycling facility management
Testing:
✅ All 22 random bucket tests passing
✅ YAML validation passing for all activities
✅ Deterministic triple-trigger test (100% probability)
Major expansion to support travel to ANY location during biblical timeline:
NEW REGIONS SUPPORTED:
- Biblical Lands: All biblical eras from Garden of Eden to persecution
- Greece: Philosophers (Socrates, Plato, Aristotle), mystery religions, gods
- Rome: Stoics, emperors, gladiators, early Christians, Roman religion
- India: Buddhist monks, Hindu gurus, yogis, karma/reincarnation
- China: Confucius, Laozi, Taoism, Confucianism, ancestor worship
- Persia: Zoroastrian magi, fire temples, dualism
- Other: Arabia, Africa, Britain, Celtic druids, etc.
KEY FEATURES:
- Geography-aware classifier: Detects both TIME and PLACE from user input
- Dynamic briefings: AI generates context for any location/time combination
- Examples: "30 AD Greece" → Athens philosophers, "500 BC India" → Buddhist monks
- NPC system supports non-biblical spiritual figures
- Conversation system respects all spiritual traditions
- Maintains Temple accuracy for biblical lands
EXAMPLES NOW WORK:
- "Take me to 30 AD Greece" → Meet Stoic philosophers
- "500 BC India" → Meet Buddha's followers
- "Moses" → Egypt ~1446 BC
- "Socrates" → Athens ~400 BC
- "Confucius" → China ~500 BC
- "Garden of Eden" → Paradise before Fall ~4000 BC
File: 662 lines (was 532), validates with 0 errors
Major changes:
- Reduced from 1608 to 532 lines (70% reduction)
- Single open-ended question: "Where/who/when would you like to visit?"
- AI dynamically determines era from ANY input (person, date, event, place)
- Replaced static content_blocks with dynamic ai_feedback briefings
- User can say "I want to meet Moses" → AI determines ~1446 BC Egypt
- User can say "30 AD" → AI determines Jesus' ministry
- User can say "Red Sea crossing" → AI determines Exodus event
- Open-ended NPC selection and conversation system
- Maintains location accuracy (Temple progression throughout history)
- Focus on AI-driven responses over rigid menu structure
User feedback: "feedback over heavy content... needs to be open ended"
Each time travel destination now includes comprehensive briefing:
**Briefing Format:**
- Destination (geographic location)
- Time Period (specific dates)
- Biblical Reference (relevant scripture)
- What You'll Experience (historical context, key events, atmosphere)
- Important/Critical Location Notes (especially Temple status)
**Educational Enhancements:**
Garden of Eden:
- Explains it's before sin, perfect creation
- Notes no buildings/cities exist yet
Fall & Early World:
- Describes life after sin entered
- Notes Cain/Abel, first altars, long lifespans
Egypt & Exodus:
- **CRITICAL:** Emphasizes NO Temple for 500+ more years
- Explains Moses uses simple altars
- Egyptian temples to Ra/Osiris present
Solomon's Temple:
- **HISTORIC MOMENT:** FIRST Temple after 480 years!
- Describes gold overlay, Ark location
- This is what Moses and David longed for
Jesus' Ministry:
- SECOND Temple (Herod's) stands
- Jesus prophesies its destruction
- Will be gone in 40 years (70 AD)
Roman Persecution:
- NO Temple (destroyed 70 AD)
- Christians meet in catacombs
- Fish symbol as secret sign
Makes Temple progression crystal clear: none → altars → First Temple → Second Temple → destroyed → underground faith.
Users now understand WHEN and WHERE they're going before arrival.
COMPLETE REWRITE with all requested features:
**Starts at the Beginning:**
- Garden of Eden (Paradise before sin)
- Fall & Early World (Cain, Abel, Enoch)
**Covers Full Bible Chronologically:**
- Egypt & Exodus (~1446 BC)
- Solomon's Temple (~970 BC)
- Life of Jesus (~30 AD)
- Persecution & Martyrdom (~64-313 AD)
**Open-Ended NPC Selection:**
- Users can request ANY biblical figure from each era
- AI dynamically rolepl ays any character accurately
- Suggestions provided but not limiting
- Examples: "Moses", "Queen of Sheba", "a Hebrew slave"
**Historically Accurate Locations:**
- Garden of Eden: NO buildings, only perfect nature
- Egypt: NO Temple to YHWH (only altars, won't exist for 500+ years)
- Solomon: FIRST Temple in all its glory
- Jesus' time: SECOND Temple (Herod's Temple)
- Persecution: NO Temple (destroyed 70 AD), catacombs instead
**Features:**
- 1608 lines, 9 sections, 25 steps
- Metadata tracking (epochs visited, people met)
- Multilingual support
- Looping time machine hub
- Biblically accurate character portrayals
- Scripture references throughout
Demonstrates location accuracy progression: no temple → Tabernacle/altars → First Temple → Second Temple → no temple (destroyed) → faith survives underground.
Create immersive time travel experience through key Biblical epochs:
- Egypt & Exodus: Meet Moses, Pharaoh, Hebrew slaves, Aaron
- Kingdom of David: Visit King David, Prophet Nathan, musicians, citizens
- Life of Jesus: Walk with Jesus, disciples, Mary Magdalene, crowds
- Pentecost & Early Church: Experience Holy Spirit, meet apostles and converts
- Roman Persecution: Stand with martyrs, Paul, persecuted believers
Features:
- Time machine hub for epoch selection
- Multiple NPCs per epoch with unique personalities
- Biblically accurate dialogue and references
- Metadata tracking for journey statistics
- Looping mechanism to revisit epochs
- Final reflection on spiritual journey
The activity maintains historical accuracy while being engaging and educational.
Create activity40-fashion-empire-backrooms.yaml with:
Features:
- Player is a girl running her own fashion brand underground
- Backrooms aesthetic: liminal warehouse spaces, mysterious locations
- 4 explorable locations: Warehouse Level -3, The Salon, Sub Bay (underwater lab), Reactor Atelier (nuclear power)
- Full control over 70+ robots and NPC employees (Zara-7, Viktor, Mx. Kai, Luna & Sol)
- Mission-based gameplay (15% tasks, 5% emergencies)
- Player makes creative, leadership, and strategic decisions
Locations:
- Warehouse Level -3: Storage backrooms, assembly drones, fabric management
- The Salon: Creative hub, style bots, runway preparation
- Sub Bay: Underwater dye laboratory, bioluminescent experiments, submersibles
- Reactor Atelier: Nuclear-powered textile synthesis, atomic fabric manipulation
Gameplay:
- Choose locations via central elevator
- Complete missions (color selection, robot commands, textile treatments, power management)
- Handle emergencies (fabric contamination crisis with multiple solutions)
- Manage NPCs and give directives
- Make creative vision decisions for runway shows
- Culminates in Neon Dreams runway show featuring player's choices
- Player sees their vision realized through their empire
All transitions validated, proper termination, educational about fashion + leadership
This activity provides an immersive, open-ended exploration of Earth's
prehistoric eras where users can:
- Travel through Triassic, Jurassic, and Cretaceous periods
- Explore climate, geography, dinosaurs, marine reptiles, and pterosaurs
- Learn about specific creatures on demand
- Understand the evolution of life and flowering plants
- Witness the K-T extinction event
- Jump freely between time periods
Features:
- Central "control room" hub for navigation
- Detailed information about 30+ dinosaurs and creatures
- Covers vegetation changes including flowering plant revolution
- Open-ended exploration with AI-guided learning
- Comprehensive extinction event explanation
- Supports looping and non-linear exploration
The activity validates successfully and follows best practices for
engagement, education, and proper termination.
Enhanced moon navigation to allow easy return to the host planet's details.
Previously, moon navigation only had "leave_jupiter" which took you to the
final navigation menu. Now you can also go back to see the planet itself.
New Navigation Pattern (implemented for Jupiter's moons):
From any moon, you can now:
1. **Jump to other moons** - "Europa", "Ganymede", etc.
2. **Return to moon menu** - "moon menu" shows all moon options
3. **Back to planet** - "Jupiter" or "back to Jupiter" returns to planet details
4. **Leave entirely** - "leave Jupiter" goes to planet-to-planet navigation
Example Navigation Flow:
- Visit Jupiter → See planet details
- Choose "Io" → See Io's volcanoes
- Say "back to Jupiter" → Return to Jupiter's details (storms, bands, etc.)
- Say "Europa" → Jump directly to Europa
- Say "leave Jupiter" → Continue to Saturn
Updated Navigation Steps:
- moon_io_nav: Added back_to_planet → jupiter:jupiter_details
- moon_europa_nav: Added back_to_planet → jupiter:jupiter_details
- moon_ganymede_nav: Added back_to_planet → jupiter:jupiter_details
- moon_callisto_nav: Added back_to_planet → jupiter:jupiter_details
- jupiter_other_moons_nav: Added back_to_planet → jupiter:jupiter_details
This same pattern can be extended to Saturn, Uranus, and Neptune moons,
allowing seamless navigation: Moon → Moon, Moon → Planet, Planet → Planet.
Enhanced the Solar System Explorer with comprehensive moon navigation menus,
allowing users to jump freely between moons within each planet's system.
New Navigation Features:
- Moon selection menus after each planet's moon intro
- Individual navigation after each moon's details
- Ability to jump directly to any moon or back to the menu
- "Stay" option to ask questions without penalty
Jupiter (95 moons):
- Moon menu: Choose Io, Europa, Ganymede, Callisto, or other moons
- Navigation after each Galilean moon
- Can jump between any moons freely
Saturn (146 moons):
- Moon menu: Choose Titan, Enceladus, Mimas, or other moons
- Navigation after each major moon
- Jump between moons or back to menu
Uranus (28 moons):
- Moon menu: Choose Miranda or other major moons
- Navigation from Miranda to other moons
- Can explore Ariel, Umbriel, Titania, Oberon via other moons
Neptune (16 moons):
- Moon menu: Choose Triton or other moons
- Navigation after Triton
- Can jump back to major moon or continue journey
Example Usage:
1. Visit Jupiter → Choose moon menu
2. Say "Europa" → See Europa details
3. Say "I want to see Io" → Jump directly to Io
4. Say "moon menu" → Back to selection
5. Say "other moons" → See smaller moons
6. Say "leave Jupiter" → Continue to Saturn
This makes exploration truly non-linear and interactive, exactly as
requested for exploring moons like jumping between Uranus's moons!
Created an interactive, open-ended educational activity that allows students to explore the entire solar system at their own pace.
Features:
- Complete coverage of all major celestial bodies
- The Sun with detailed structure and solar activity
- All 8 planets with comprehensive details
- 200+ moons documented with attributes:
- Jupiter's 95 moons (4 Galilean moons + others)
- Saturn's 146 moons (Titan, Enceladus, Mimas, etc.)
- Uranus's 28 moons (Miranda and major moons)
- Neptune's 16 moons (Triton and others)
- Earth's Moon, Mars's Phobos & Deimos
- Asteroid Belt (Ceres, Vesta, Pallas, Hygiea)
- Kuiper Belt (Pluto, Eris, Makemake, Haumea, etc.)
Activity Structure:
- 13 main sections (Introduction, Sun, 8 planets, Asteroid Belt, Kuiper Belt, Conclusion)
- Non-linear exploration - jump to any location at any time
- Detailed scientific information with current data
- Engaging presentation with emojis and formatting
- Educational content based on latest discoveries (New Horizons, Cassini, Juno missions)
Technical:
- Fully validated YAML structure
- All transitions properly mapped
- Proper termination paths
- Interactive Q&A at each location
- "Stay" option allows asking questions without counting as attempts
Perfect for astronomy education and space exploration learning!
NEW ACTIVITIES:
activity48-monty-hall-simulation.yaml - Monty Hall paradox proof
- Simulate stay vs switch strategies
- Prove switching wins 2/3 through code
- Any programming language support
activity49-multi-armed-bandit.yaml - Adaptive algorithms beat A/B testing
- Epsilon-greedy implementation
- 88% regret reduction vs traditional A/B
- Real-world applications (web optimization, clinical trials)
activity50-genetic-algorithms.yaml - Evolution-based optimization
- String evolution challenge
- Fitness, selection, crossover, mutation
- 803,181x faster than brute force
activity51-connect-four.yaml - Complete game development
- 2D arrays and game state
- Win detection algorithms (horizontal, vertical, diagonal)
- Full game loop implementation
All activities:
- Support ANY programming language choice
- Follow pedagogical best practices (concepts first, code in feedback)
- Validate with zero errors/warnings
- Engaging and fun (aha moments, real games, simulations)
NEW ACTIVITIES:
activity46-game-theory-python.yaml - Game theory implementation in Python
- Representing games with dictionaries
- Payoff matrix as dict with tuple keys
- Query functions and game simulation
- One-shot and repeated games
- Tit-for-Tat strategy implementation
- Function composition and abstraction
activity47-game-theory-c.yaml - Game theory implementation in C
- Defining Payoff struct for outcomes
- 2D arrays for payoff matrices
- Memory-efficient game representation
- Strategy lookup functions
- Enum for self-documenting code
- Pointer and struct fundamentals
Both activities:
- Teach programming through game theory concepts
- Follow pedagogical best practice (concepts first, code examples in feedback)
- Validate with zero errors/warnings
- Progressive difficulty (structures → functions → simulation)
- Real-world application of abstract concepts
- Engage students with strategic thinking + coding
PROBLEM: activity37 was showing complete code examples in Python, JavaScript,
Java, and C++ BEFORE asking students to write code themselves. This turns
learning into copy-paste practice.
FIXED:
- Hello World section: Removed multi-language code examples from content_blocks
- Variables section: Removed multi-language code examples from content_blocks
- Now explains CONCEPTS (what, why, how languages differ) without showing syntax
- Code examples remain in AI feedback for when students struggle or ask for help
PEDAGOGICAL APPROACH:
1. Explain the concept (stdout, variables, etc.)
2. Explain language differences conceptually (dynamic vs static typing)
3. Ask students to TRY writing code in THEIR language
4. Provide language-specific examples in AI FEEDBACK if they struggle
This way students actually have to THINK and LEARN, not just copy.
UPDATED CLAUDE.md:
- Added new pitfall: "Showing answers before questions"
- Guidance: Explain concepts in content_blocks, provide code examples in ai_feedback
Still validates perfectly with zero errors/warnings.
activity38-fashion-today.yaml - Fun, interactive style discovery
- Personal style identification (classic, boho, streetwear, etc.)
- Color psychology and preferences
- Outfit building for occasions
- Statement pieces and accessories
- Fashion philosophy reflection
- Encourages self-expression and confidence
activity39-fashion-history.yaml - Educational timeline 1800-2025
- Victorian era corsets and social restrictions
- 1920s flappers and women's liberation
- WWII rationing and practical fashion
- 1950s ultra-femininity and gender politics
- 1960s-70s revolution (mod, hippie, disco, punk)
- 1980s excess and 1990s grunge backlash
- 2000s-2010s fast fashion and social media
- 2020s sustainability, inclusivity, technology
- Critical thinking about fashion as social mirror
Both activities:
- Follow expert guide validation requirements
- Include engaging content with emojis and formatting
- Support language switching
- Use metadata strategically
- Have multiple response paths with tailored feedback
- Terminate properly with activity_completed markers
- Passed activity_yaml_validator.py with zero errors/warnings
- Increased from 1,148 to 1,895 lines (+65%)
- Added detailed explanations before each coding exercise
- Enhanced Hello World section with stdout concepts and multi-language examples
- Expanded Variables section with box analogy, naming rules, and typing differences
- Enhanced Data Types with comprehensive type explanations and string formatting
- Expanded If Statements with conditional logic fundamentals and comparison operators
- Enhanced Loops with detailed for loop explanations, execution traces, and common patterns
- Expanded Functions with DRY principle, parameter explanations, and best practices
- Enhanced Return Values with display vs return differences and common mistakes
- All sections now teach fundamentals thoroughly before asking students to code
- Validation passed successfully
This commit addresses two critical issues:
1. Completion Bug (activities 30-37):
- Final steps were looping forever, preventing activity completion
- Fixed by removing next_section_and_step from completion transitions
- Kept off_topic transition looping to avoid validator terminal step errors
- Activities now complete properly when users give valid final answers
2. Activity37 Bucket Logic:
- Changed "close" bucket to retry same step instead of advancing
- Only "correct" bucket now advances to next step
- All other buckets (close, incomplete, wrong_language, etc.) retry
- This ensures students must get correct answers to progress
Technical Details:
- Final steps are not considered "terminal" if at least one transition
has next_section_and_step (validator requirement)
- Off-topic transitions loop back to allow another attempt
- Completion happens when get_next_step() returns None, None
Validation:
- All 8 activities pass activity_yaml_validator.py
- No errors or warnings
Affects: activity30-37 (all new merged activities)
When MODEL_X references (MODEL_0, MODEL_1, etc.) are used, the code now
properly looks up actual model names from the dynamic registry (MODEL_CLIENT_MAP)
instead of hardcoding "model" or requiring MODEL_NAME_X environment variables.
Changes:
- app.py: Look up models from MODEL_CLIENT_MAP for the specified endpoint
- guarded_ai.py: Query endpoints for actual model names at initialization
- guarded_ai.py: Use dynamic registry for MODEL_X lookups
This fixes the "model not found" error when using activities with MODEL_X
references like activity37.
Respects existing stable configuration where:
- MODEL_0 = Hermes (default for classification and feedback)
- MODEL_1 = Qwen (for code generation)
- MODEL_2 = GPT
Updated:
- All function defaults in activity.py: MODEL_1 -> MODEL_0
- activity37: Uses MODEL_0 for classification, MODEL_1 for code feedback
This works with the existing environment variable setup without requiring changes to vars.sh.
Critical fix for guarded_ai.py:
- Add MODEL_NAME_{n} environment variable support
- Fixes hard-coded "model" string that breaks Azure OpenAI and other endpoints
- Falls back to "model" if MODEL_NAME_{n} not specified
- Some endpoints require actual deployment name in model parameter
Validator improvement:
- Allow feedback_prompts as alternative to feedback_tokens_for_ai
- Prevents false warning when using metadata_feedback_filter with new prompt system
Documentation:
- Added MODEL_NAME_{n} examples to CLAUDE.md
- Documented that Azure and similar endpoints need this variable
All 8 activities validated: 0 errors, 0 warnings
Transformed planning document into comprehensive completion report:
- Status: 8 activities completed (30-37), 6,112 lines of YAML
- Documented new classifier_model and feedback_model feature
- Added model setup guide for Qwen3-Coder-30B
- Detailed activity summaries with special features
- Technical architecture and implementation decisions
- Usage examples and future enhancements
Key highlights:
- All activities validated with 0 errors
- Dual-model architecture explained
- Activity 37 flagship feature: universal programming language support
- Hermes excellence in role-playing scenarios
Added detailed comments showing how to use the recommended model:
- hf.co/unsloth/Qwen3-Coder-30B-A3B-Instruct-GGUF:Q4_K_M
- Setup instructions for llama.cpp (with GPU offloading)
- Alternative setup with ollama
- Environment variable configuration examples
This 30B parameter model is specifically optimized for code generation
across all programming languages, making it perfect for the universal
programming activity.
Changes:
- Enhanced get_openai_client_and_model() to support MODEL_X references
- Added model parameter (default "MODEL_1") to all AI functions:
- categorize_response()
- generate_ai_feedback()
- provide_feedback()
- provide_feedback_prompts()
- translate_text()
- Updated simulate_activity() to:
- Read classifier_model and feedback_model from YAML
- Support step-level model overrides
- Pass appropriate models to classifier vs feedback functions
This ensures the CLI simulation tool matches the production activity.py behavior.
Allow activities to specify separate models for classification and feedback:
- classifier_model: Used for categorizing user responses into buckets
- feedback_model: Used for generating AI feedback and translations
Both fields can be set at activity level (defaults) and overridden at step level.
Updated activity37 to use:
- MODEL_1 (Hermes) for classification
- MODEL_3 (Qwen 3 Coder) for feedback
This allows using specialized models for different tasks, e.g., fast classification
with accurate feedback generation from domain-specific models.
Created 3 comprehensive educational activities without embedded Python:
1. activity35-american-history.yaml - Advanced American History for gifted students
- Founding principles and Constitutional design
- Civil War causes and Reconstruction failure
- Civil Rights Movement strategies
- Primary source analysis and critical historical thinking
- Connects past to present issues
2. activity36-biblical-history.yaml - Biblical History & Ancient Near East
- Ancient Near Eastern context (Mesopotamia, Egypt, Canaan)
- Archaeological evidence and historical reconstruction
- Israelite history (Exodus, Monarchy, Exile)
- Roman period and early Christianity
- Foundation myths vs historical facts
- Cultural adaptation and religious transformation
3. activity37-programming-languages.yaml - Universal Programming Concepts
- Student chooses ANY programming language (Python, C++, COBOL, anything)
- AI adapts all examples/feedback to chosen language via metadata
- Covers: stdout/output, variables, data types, control flow, loops, functions
- All examples use stdout to display messages
- Concepts applicable to every language
- Language-specific syntax provided by AI
All activities:
- Use only YAML features (no embedded Python)
- Validate successfully with 0 errors
- Provide sophisticated educational content
- Use AI feedback for personalization
- Include critical thinking and reflection
- Track progress via metadata
Total: 8 new educational activities across 2 commits (5 from previous commit + 3 now)
Created 5 comprehensive educational activities that use only YAML features
(buckets, transitions, metadata operations, AI feedback) without Python scripts:
- activity30-logic-puzzles.yaml: Critical thinking through deductive reasoning,
contrapositives, syllogisms, and knights/knaves puzzles
- activity31-scientific-method.yaml: Learn scientific method through historical
case studies (Semmelweis, Newton) and experimental design principles
- activity32-world-geography.yaml: Choose-your-own-adventure journey exploring
continents, countries, capitals, and cultural facts
- activity33-environmental-science.yaml: Role-playing as environmental consultant
making sustainability decisions on transportation, energy, land use, waste, and food
- activity34-media-literacy.yaml: Develop critical media consumption skills,
evaluate sources, recognize bias, fact-check claims, and spot manipulation
All activities:
- Follow existing YAML schema and validate successfully
- Use Socratic buckets for educational feedback
- Include set_language support
- Track progress via metadata
- Provide AI-generated personalized feedback
- Are educational, engaging, and progressively challenging
- Include final reflection steps
Also added NEW_ACTIVITIES_PLAN.md documenting the planning process and
design rationale for each activity.
- Remove duplicate messages from section transitions like activity14
- Fix coin categorization issue - 'use coin' was being misclassified as 'use_key_and_password'
- Add section_4:step_2 for post-safe-opening state with proper coin slot options
- Update tokens_for_ai to properly distinguish between different user actions
- Now players can properly access the secret compartment using the coin
- Activity validated and passes all checks
Remove prescriptive ship destruction descriptions and let the AI be creative.
Since skip_condition ensures these prompts only run when ships are actually
destroyed, we can make the prompts more concise and focused on the outcome.
Add skip_condition logic to feedback prompts to prevent AI from generating
false ship destruction messages when no ships were actually destroyed.
Changes:
- Add skip_condition parameter support in provide_feedback_prompts()
- Support all_null, all_false, and all_true condition types
- Apply skip_condition to battleship Ship Status and Game Over prompts
- Add comprehensive unit tests covering all skip condition scenarios
- Test real battleship scenario that was causing hallucinations
This prevents the AI from creating false positive ship destruction messages
when metadata indicates no ships were actually sunk (all null values).
- Removed feedback_tokens_for_ai from step 3 Game Over
- Exit transition already has appropriate content_blocks
- Eliminates duplicate farewell messages when exiting
Major improvements to battleship game feedback accuracy and user experience:
## New Multi-Prompt Feedback System
- Replaced single feedback with 3 specialized prompts: Shot Report, Ship Status, Game Over
- Each prompt has individual metadata filtering to see only relevant data
- Shot Report only sees hit/miss data, Ship Status only sees ship destruction data
- Added STFU token system to suppress empty messages (filtered out automatically)
## Technical Implementation
- Added per-prompt metadata_filter support in YAML structure
- Updated app.py and guarded_ai.py to handle prompt-specific filtering
- Legacy single-prompt system still works with transition-level filtering
- Added comprehensive test suite for feedback system validation
## User Experience Fixes
- Fixed TTS queue blocking JavaScript execution (async promises instead of await)
- Ship Status now correctly reports who destroyed which ship (role confusion fixed)
- Game Over only appears when game actually ends (no more random messages)
- Maintained dramatic storytelling while ensuring factual accuracy
## Battleship-Specific Improvements
- Ship destruction messages only appear when ships actually sink
- Clear separation of concerns: hits/misses vs ship destruction vs game over
- Eliminated false positive ship destruction reports
- Fixed role reversal where wrong player got credit for destruction
The battleship narrator now provides accurate, contextual feedback while preserving the dramatic naval warfare atmosphere.
- Fix battleship feedback perspective confusion with better Hermes prompting
- Add auto-play TTS button with localStorage persistence and queueing system
- Move activity controls below model/voice selectors in sidebar
- Add activity controls to mobile hamburger menu
- Fix model/activity dropdowns to stay within container bounds
- Filter activities API to only show .yaml/.yml files
- Clean up system message labels by moving to usernames (System (Feedback), System (Question))
- Apply black formatting to app.py
- Updated validator terminal step detection to only flag truly terminal steps
- Fixed validator to accept integers and booleans in buckets (as supported by app.py)
- Fixed metadata_remove format in activity17 from dictionary to list of strings
- Added proper terminal section to activity3.yaml without questions/buckets
- Fixed missing restart transition and bucket in activity28
- Removed unused game_end transitions from battleship files
- Updated exit transitions to go directly to step_4 (goodbye step)
- Applied black formatting to validator code
All 30 activity YAML files now validate successfully with 0 errors and 0 warnings.
- Add matplotlib.use("Agg") backend configuration to prevent runtime errors in headless environments
- Add error handling guards for script results that might return None
- Fix AI targeting logic to exclude already-fired cells in super hunter and hunter modes
- Update CLAUDE.md with matplotlib best practices
- Add pass statements to empty else blocks that only contained commented prints
- Ensures Python syntax remains valid after commenting out debug statements
- Add user_response to pre-script metadata for better game state management
- Implement metadata_feedback_filter to control feedback data exposure
- Improve ship destruction announcements and game over messaging
- Add debug logging for ship sinking events
- Include test ship configuration file
- Add new Hermes Reasoner AI mode that combines probability analysis with LLM reasoning
- Implement pre-script and post-script architecture in app.py for flexible YAML processing
- Fix game ending detection by adding transition override mechanism
- Add probability matrix visualization and strategic move analysis
- Support both legacy processing_script and new pre_script/post_script naming
- Restore full ship complement for complete battleship gameplay
- Replace unsafe eval() with sympy for secure expression parsing
- Add YAML anchors to eliminate code duplication in processing scripts
- Implement multiple function plotting with comma-separated syntax
- Add dynamic plot ranges based on function characteristics
- Include automatic function type detection and analysis
- Streamline activity flow: intro → demo plot → open sandbox
- Add comprehensive error handling with visual error messages
- Support enhanced mathematical notation (arcsin, ln, implied multiplication)
modified: research/activity24-math-plot.yaml
