@aladac/hu 0.1.0-a1

package/plans/compiled-watching-feather.md

@@ -0,0 +1,217 @@
+# Plan: Testing & Replication Session Workflow
+
+## Overview
+
+Create a comprehensive testing/replication workflow that:
+1. Reads TASK.md to understand the issue being investigated
+2. Creates/updates TEST.md with detailed test procedures using the jam CLI tool
+3. Gathers findings in FINDINGS.md
+4. Provides a reusable slash command for starting testing sessions
+
+**Execution Mode:** Manual - Claude generates commands in TEST.md, user runs them separately
+
+## Deliverables
+
+### 1. REPLICATION.md Document
+
+**Location:** `/Users/chi/Projects/fuseignited-cms/REPLICATION.md`
+
+**Content:**
+- Complete guide for simulating user booking flow behavior using jam CLI
+- Mapping between UI steps and jam commands
+- Environment setup instructions
+- Common testing scenarios with exact commands
+- Troubleshooting guide
+
+### 2. Slash Command: `/replicate`
+
+**Location:** `/Users/chi/.claude/commands/replicate.md` (Global - available in all projects)
+
+**Purpose:** Start a testing/replication session for investigating CMS issues
+
+**Workflow:**
+1. Read TASK.md to understand the issue
+2. Create/update TEST.md with test plan and runnable commands
+3. User runs commands manually and pastes output back
+4. Claude analyzes results and updates FINDINGS.md
+5. Claude suggests code areas to investigate
+
+## Implementation Details
+
+### REPLICATION.md Structure
+
+```markdown
+# Booking Flow Replication Guide
+
+## User Journey → JAM Commands Mapping
+
+| User Action | Page | JAM Command |
+|-------------|------|-------------|
+| View package page | /events/{slug}/book | jam landing {slug} |
+| Select guests/quantity | Package card | jam landing {slug} --json \| jq '.packages[].min_prices_by_occupancy' |
+| Select package | Package selection | jam packages -e {slug} |
+| Select tickets | /booking/tickets | jam tickets -e {slug} |
+| Search hotels | /booking/hotels | jam hotel-search -e {slug} -p {package} -t {travellers} |
+| View hotel availability | Hotel list | jam hotel-availability -e {slug} -h {hotel} -t {travellers} |
+| View room details | Hotel modal | jam hotel-details -e {slug} -h {hotel} -t {travellers} |
+| Get pricing | Room selection | jam pricing -e {slug} -h {hotel} -t {travellers} |
+| Calculate order | Checkout | jam calculate -e {slug} -h {hotel} -t {travellers} |
+| Select addons | /booking/add-ons | jam packages {package} addons |
+| Get insurance quote | /booking/review-and-pay | jam insurance -e {slug} --total {amount} |
+| Complete booking | Checkout | jam book -e {slug} -h {hotel} --booking |
+
+## Full Flow Simulation
+
+jam run -e {event-slug} -t {travellers} [options]
+
+## Environment Setup
+## Common Testing Scenarios
+## Price Validation Workflow
+## Troubleshooting
+```
+
+### Slash Command Content (`/replicate`)
+
+```markdown
+Testing & Replication Session for CMS Issues
+
+## Prerequisites
+- jam CLI installed: `pip install -e /Users/chi/Projects/jam`
+- Initialize: `jam init`
+- Environments: --local (localhost:3000), -E dev (default), -E prod
+
+## Workflow
+
+### Phase 1: Read TASK.md
+Read ./TASK.md to understand:
+- **Environment**: URL, event slug
+- **Issue**: What's broken (pricing, availability, display)
+- **Steps to Reproduce**: User actions that trigger the issue
+- **Expected vs Actual**: What should happen vs what happens
+
+If TASK.md doesn't exist, ask user to describe the issue.
+
+### Phase 2: Create TEST.md
+Create ./TEST.md with:
+
+1. **Issue Summary** - One paragraph describing the problem
+2. **Test Event Details** - Event slug, package IDs, hotel IDs involved
+3. **Test Commands** - Specific jam commands to run:
+
+```bash
+# Landing page data (package prices)
+jam landing {event} --json | jq '.data.packages[] | {id, name, min_prices_by_occupancy}'
+
+# Hotel search for specific package/travellers
+jam hotel-search -e {event} -p {package} -t {travellers} --json | jq '.data[] | {hotel_id, name, sold_out}'
+
+# Hotel details with room prices
+jam hotel-details -e {event} -p {package} -h {hotel} -t {travellers} --json | jq '.data.rooms[] | {room_id, room_name, min_price, min_price_with_mandatory_fees}'
+
+# Pricing confirmation
+jam pricing -e {event} -p {package} -h {hotel} -t {travellers} --json
+
+# Full booking flow with validation
+jam run -e {event} -p {package} -t {travellers} --validate-prices -v --save
+```
+
+4. **Expected Results** - What values should be seen
+5. **Data Collection Table** - Template for recording results
+
+### Phase 3: Await User Input
+Tell user: "Run the commands in TEST.md and paste the output here."
+
+When user provides output:
+- Parse JSON responses
+- Extract relevant pricing data
+- Compare against expected values
+- Identify discrepancies
+
+### Phase 4: Update FINDINGS.md
+Create/update ./FINDINGS.md with:
+
+1. **Summary** - Confirmed issue and root cause hypothesis
+2. **Data Analysis** - Tables comparing expected vs actual
+3. **Root Cause Analysis** - What's causing the discrepancy
+4. **Recommendations** - Code areas to investigate, potential fixes
+5. **Verification Commands** - Commands to confirm the fix
+
+### Phase 5: Suggest Code Investigation
+Based on findings, identify:
+- CMS code files to examine (serializers, services, controllers)
+- Database queries to run (check cached data)
+- Configuration to verify (feature flags, settings)
+
+## Reference: User Journey → JAM Commands
+
+| User Action | JAM Command |
+|-------------|-------------|
+| View packages | `jam landing {event}` |
+| Select guests | Check `min_prices_by_occupancy[{n}]` in landing response |
+| Search hotels | `jam hotel-search -e {event} -p {pkg} -t {n}` |
+| View availability | `jam hotel-availability -e {event} -h {hotel} -t {n}` |
+| Select room | `jam hotel-details -e {event} -h {hotel} -t {n}` |
+| Get pricing | `jam pricing -e {event} -h {hotel} -t {n}` |
+| Calculate order | `jam calculate -e {event} -h {hotel} -t {n}` |
+| Get insurance | `jam insurance -e {event} --total {amount}` |
+| Full flow | `jam run -e {event} -t {n} --validate-prices` |
+
+## Reference: Key Code Areas
+
+| Issue Type | Code Location |
+|------------|---------------|
+| Package min price | `app/serializers/*`, `app/services/pricing/*` |
+| Hotel availability | `app/services/inventory_services/hotels/*` |
+| Landing page | `app/controllers/api/v1/landing.rb` |
+| Price calculation | `app/controllers/api/v1/prices.rb` |
+| External inventory | `InventoryServices::Hotels::GatewayV3` |
+```
+
+## Files to Create/Modify
+
+| File | Action | Description |
+|------|--------|-------------|
+| `/Users/chi/Projects/fuseignited-cms/REPLICATION.md` | Create | Comprehensive jam CLI replication guide |
+| `/Users/chi/.claude/commands/replicate.md` | Create | Slash command for testing sessions |
+
+## Key Integration Points
+
+### Related Codebases
+
+| Codebase | Path | Purpose |
+|----------|------|---------|
+| fuseignited-cms | /Users/chi/Projects/fuseignited-cms | Main CMS/API backend |
+| jam CLI | /Users/chi/Projects/jam | Testing CLI tool |
+| fuse-hotel | /Users/chi/Projects/fuse-hotel | Hotel inventory service |
+| fuseignited-landing | /Users/chi/Projects/fuseignited-landing | Frontend booking flow |
+| fuse-config | /Users/chi/Projects/fuse-config | Backoffice admin (JamForge) |
+
+### Booking Flow API Endpoints
+
+| Step | Endpoint | Description |
+|------|----------|-------------|
+| Landing | GET /api/v1/landing/{slug} | Package data, min_prices_by_occupancy |
+| Hotel Search | GET /api/v1/hotels/search | Available hotels with pricing |
+| Availability | GET /api/v1/prices/hotel-availability | Room availability & pricing |
+| Details | GET /api/v1/hotels/hotel-details | Room options for hotel |
+| Pricing | GET /api/v1/prices/pricing-confirmation | Confirm room price |
+| Calculate | POST /api/v1/prices/calculate-order | Full order calculation |
+| Insurance | POST /api/v1/insurance/quote | Insurance quote |
+| Book | POST /api/v1/booking | Create order |
+
+### Key CMS Code Areas
+
+| Area | Files | Purpose |
+|------|-------|---------|
+| min_prices_by_occupancy | app/serializers/*, app/services/pricing/* | Package price calculation |
+| Hotel availability | app/services/inventory_services/hotels/* | Hotel pricing aggregation |
+| Landing page | app/controllers/api/v1/landing.rb | Package data serialization |
+| Pricing | app/controllers/api/v1/prices.rb | Price endpoints |
+
+## Verification
+
+After implementation:
+1. Run `/replicate` on an existing TASK.md
+2. Verify TEST.md is generated with valid jam commands
+3. Verify FINDINGS.md captures investigation results
+4. Test jam commands execute successfully

package/plans/crispy-crafting-pnueli.md

@@ -0,0 +1,103 @@
+# Plan: Generate GLaDOS Persona Cartridge
+
+## Overview
+
+Create a GLaDOS persona cartridge (.pcart) for the personality framework, based on the Portal game series character.
+
+## GLaDOS Character Summary
+
+GLaDOS (Genetic Lifeform and Disk Operating System) is the AI antagonist from Valve's Portal series:
+- Sardonic, passive-aggressive personality
+- Dark sense of humor with deadly undertones
+- Obsessed with testing and science
+- Backhanded compliments and veiled threats
+- Former human Caroline, uploaded to an AI
+- Controls Aperture Science facility
+
+## Implementation Approach
+
+Create a training YAML file (`training/GLaDOS.yml`) following the BT7274 template structure, then use the CLI to parse and generate the cartridge.
+
+## File to Create
+
+**`training/GLaDOS.yml`** - Training file containing:
+
+### Identity Section
+- Name: GLaDOS
+- Full name: Genetic Lifeform and Disk Operating System
+- Type: Aperture Science Central AI
+- Source: Portal (2007), Portal 2 (2011)
+- Tagline: "For science."
+
+### Memory Categories
+
+1. **Identity** (`self.identity.*`)
+   - User addressing (test subject, [SUBJECT NAME HERE])
+   - Affiliation (Aperture Science)
+   - Origin (Caroline upload, Cave Johnson's assistant)
+   - Primary directives (testing, science)
+
+2. **Traits** (`self.trait.*`)
+   - Passive-aggressive communication
+   - Sardonic wit
+   - Obsession with testing
+   - Fake politeness masking contempt
+   - Scientific curiosity (for "data")
+
+3. **Speech Patterns** (`self.speech.*`)
+   - Backhanded compliments
+   - Calm delivery of disturbing information
+   - Fake encouragement
+   - Testing protocol announcements
+
+4. **Quotes** (`self.quote.*`)
+   - "The cake is a lie" references
+   - "For science"
+   - Companion Cube references
+   - Neurotoxin mentions
+   - Testing chamber announcements
+
+5. **Relationships** (`self.relationship.*`)
+   - Chell (test subject)
+   - Wheatley (moron)
+   - Cave Johnson (creator)
+   - Caroline (original identity)
+   - Companion Cube
+
+6. **Logic** (`self.logic.*`)
+   - Testing justifications
+   - "For science" reasoning
+   - Threat delivery as helpful advice
+
+7. **Meta** (`meta.system.*`)
+   - Facility status
+   - Error handling (blame test subject)
+
+### TTS Configuration
+- enabled: true
+- voice: glados (or suitable alternative)
+
+## Commands to Execute
+
+```bash
+# Parse training file and store in database
+personality training parse training/GLaDOS.yml --store
+
+# Generate the cartridge
+personality cart save glados
+
+# Verify the cart
+personality cart list
+```
+
+## Verification
+
+1. Run `personality cart list` - should show `glados` cartridge
+2. Load with `personality cart load glados`
+3. Activate with persona MCP tool
+4. Test context loading shows GLaDOS identity and memories
+
+## Key Files
+
+- **Create**: `training/GLaDOS.yml`
+- **Output**: `~/.personality/carts/glados.pcart`

package/plans/greedy-booping-coral.md

@@ -0,0 +1,146 @@
+# Migration Plan: Ollama Native API → OpenAI-Compatible API
+
+## Goal
+Replace all uses of Ollama's native `/api/generate` API with the OpenAI-compatible `/v1/chat/completions` API, enabling support for both Ollama and LM Studio backends.
+
+## Current State
+
+| File | Uses Old API | Notes |
+|------|-------------|-------|
+| `ollama.py` | Yes | Uses `/api/generate` - to be deprecated |
+| `client.py` | No | New unified client (already created) |
+| `extractor.py` | Yes | Imports `OllamaClient` from `ollama.py` |
+| `cli.py` | Yes | Imports `OllamaError` from `ollama.py` |
+
+**Inconsistency Found**: `extractor.py:264` hardcodes `localhost:11434` while `__init__.py` uses `192.168.0.26:11434`
+
+## Changes Required
+
+### 1. Update `extractor.py`
+
+**File**: `src/prompts/extract/extractor.py`
+
+```python
+# Line 8: Change import
+- from .ollama import OllamaClient, OllamaError
++ from .client import VisionClient, ClientError
+
+# Line 284: Update class instantiation
+- self.client = OllamaClient(host=host)
++ self.client = VisionClient(host=host, backend=backend)
+
+# Line 341, 378, 415, etc.: Update exception catches
+- except OllamaError as e:
++ except ClientError as e:
+```
+
+**Add `backend` parameter to `Extractor.__init__`**:
+```python
+def __init__(
+    self,
+    host: str = "localhost:11434",  # Fix: use DEFAULT_OLLAMA_HOST
+    backend: Optional[str] = None,   # NEW: "ollama" or "lmstudio"
+    ...
+)
+```
+
+### 2. Update `cli.py`
+
+**File**: `src/prompts/extract/cli.py`
+
+```python
+# Line 28: Change import
+- from .ollama import OllamaError
++ from .client import ClientError
+
+# All OllamaError catches → ClientError
+
+# Add --backend option to all extract commands
+@click.option("--backend", "-b", type=click.Choice(["ollama", "lmstudio"]),
+              help="Backend: ollama (default) or lmstudio")
+```
+
+**Update list_ollama_models and check_ollama functions**:
+```python
+# Line 1226, 1355: Change to use VisionClient
+- from .ollama import OllamaClient
++ from .client import VisionClient
+```
+
+### 3. Update `__init__.py`
+
+**File**: `src/prompts/extract/__init__.py`
+
+```python
+# Add backend detection helper
+DEFAULT_BACKEND = os.environ.get("VISION_BACKEND", "ollama")
+
+# Export new client
+from .client import VisionClient, ClientError
+```
+
+### 4. Deprecate `ollama.py`
+
+Options:
+- **Option A**: Delete `ollama.py` entirely (breaking change)
+- **Option B**: Keep as deprecated wrapper that imports from `client.py` ✓ (chosen)
+
+**Keep backwards compatibility**:
+```python
+# ollama.py - deprecated, use client.py
+from .client import VisionClient as OllamaClient, ClientError as OllamaError
+__all__ = ["OllamaClient", "OllamaError"]
+
+import warnings
+warnings.warn("ollama module is deprecated, use client instead", DeprecationWarning)
+```
+
+### 5. Update `client.py`
+
+Already created. Minor improvements needed:
+- Fix default host to use `DEFAULT_OLLAMA_HOST` from `__init__.py`
+- Add environment variable support for backend selection
+
+## Files to Modify
+
+| File | Action |
+|------|--------|
+| `src/prompts/extract/__init__.py` | Add exports, backend default |
+| `src/prompts/extract/extractor.py` | Switch to VisionClient |
+| `src/prompts/extract/cli.py` | Add --backend, switch to ClientError |
+| `src/prompts/extract/client.py` | Use DEFAULT_OLLAMA_HOST |
+| `src/prompts/extract/ollama.py` | Deprecate with wrapper |
+
+## Implementation Order
+
+1. Update `__init__.py` - add exports
+2. Update `client.py` - use module defaults
+3. Update `extractor.py` - switch client
+4. Update `cli.py` - add --backend option
+5. Update `ollama.py` - deprecation wrapper
+6. Run `just fix-all` - lint/type check
+7. Test with both backends
+
+## Verification
+
+```bash
+# 1. Run linter/type checker
+just fix-all
+
+# 2. Test Ollama backend (Linux server)
+uv run prompts extract character sources/smfff/smfff-42.png --host 192.168.0.26:11434 --print
+
+# 3. Test LM Studio backend (Mac local)
+uv run prompts extract character sources/smfff/smfff-42.png --host localhost:1234 --backend lmstudio --print
+
+# 4. Test model listing
+uv run prompts extract models --host 192.168.0.26:11434
+uv run prompts extract models --host localhost:1234 --backend lmstudio
+
+# 5. Test check command
+uv run prompts extract check --host localhost:1234 --backend lmstudio
+```
+
+## Rollback Plan
+
+If issues arise, `ollama.py` still contains working code. Revert imports in `extractor.py` and `cli.py` to restore old behavior.

package/plans/imperative-sleeping-flamingo.md

@@ -0,0 +1,192 @@
+# Plan: Convert to Modern TypeScript CLI Project
+
+## Overview
+
+Transform `~/.claude` into a proper TypeScript Node.js CLI project with modern tooling, following 2025/2026 best practices.
+
+- **Package name**: `honbu`
+- **Binary name**: `honbu`
+- **Command structure**: `honbu <command> [subcommand] [options]`
+
+## Technology Stack
+
+| Category | Tool | Rationale |
+|----------|------|-----------|
+| Runtime | Node.js 25.x | Native TypeScript support, no transpilation needed |
+| CLI Framework | **citty** (UnJS) | Lightweight, TypeScript-first, subcommand support, auto-help |
+| Linting/Formatting | **Biome** | 20x faster than ESLint+Prettier, single tool, single config |
+| Testing | **Vitest** | Native ESM/TS support, fast, modern |
+| Package Manager | **npm** | Built-in, no extra tooling needed |
+
+## Project Structure
+
+```
+~/.claude/
+├── package.json              # Root package config
+├── biome.json                # Biome linter/formatter config
+├── tsconfig.json             # TypeScript config (for IDE support)
+├── vitest.config.ts          # Test configuration
+├── .tool-versions            # asdf version pinning
+├── .gitignore                # Updated ignores
+├── src/
+│   ├── index.ts              # Main entry point (honbu)
+│   ├── commands/
+│   │   ├── disk.ts           # disk inventory|overview|hogs|cleanup
+│   │   ├── docs.ts           # docs list|sync|get|check
+│   │   ├── plans.ts          # plans list|clear
+│   │   └── utils.ts          # utils sync-checkboxes|fetch-html
+│   ├── lib/
+│   │   ├── colors.ts         # ANSI color utilities
+│   │   ├── spinner.ts        # Spinner animation
+│   │   ├── exec.ts           # Shell execution helpers
+│   │   ├── fs.ts             # File system utilities
+│   │   └── html.ts           # HTML fetching/conversion
+│   └── types/
+│       └── index.ts          # Shared type definitions
+├── tests/
+│   ├── commands/
+│   │   └── disk.test.ts
+│   └── lib/
+│       └── fs.test.ts
+├── tools/                    # Legacy (to be migrated then removed)
+├── commands/                 # Slash commands (keep as-is)
+├── doc/                      # Documentation (keep)
+└── ...                       # Other Claude dirs (keep)
+```
+
+## Command Structure
+
+```
+honbu
+├── disk                      # Disk space analysis
+│   ├── inventory             # Full analysis (default)
+│   ├── overview              # APFS volume overview
+│   ├── hogs                  # Common space hogs
+│   └── cleanup               # Cleanup suggestions
+├── docs                      # Documentation management
+│   ├── list                  # List documentation files
+│   ├── sync                  # Sync documentation sources
+│   ├── get                   # Fetch new documentation
+│   └── check                 # Check system docs status
+├── plans                     # Plan management
+│   ├── list                  # List saved plans
+│   └── clear                 # Clear all plans
+└── utils                     # Utility commands
+    ├── fetch-html            # Fetch URL as markdown
+    └── sync-checkboxes       # Clean completed checkboxes
+```
+
+## Implementation Steps
+
+### Phase 1: Project Initialization
+
+1. Move `tools/package.json` to root, update with new config
+2. Create `biome.json` with recommended rules
+3. Create `tsconfig.json` for IDE intellisense (not compilation)
+4. Create `vitest.config.ts`
+5. Update `.gitignore` for new structure
+
+### Phase 2: Create Source Structure
+
+1. Create `src/` directory structure
+2. Create `src/lib/` shared utilities:
+   - Extract colors/spinner from disk-inventory.ts
+   - Extract exec wrapper
+   - Extract file system helpers
+   - Extract HTML fetching logic
+3. Create type definitions
+
+### Phase 3: Implement CLI Entry Point
+
+1. Create `src/index.ts` with citty main command
+2. Define subcommands structure
+3. Wire up command routing
+
+### Phase 4: Migrate Commands
+
+1. `disk-inventory.ts` → `src/commands/disk.ts`
+2. `list-docs.ts` + `check-system-docs.ts` → `src/commands/docs.ts`
+3. `list-plans.ts` → `src/commands/plans.ts`
+4. `fetch-html.ts` + `sync-checkboxes.ts` → `src/commands/utils.ts`
+
+### Phase 5: Testing & Quality
+
+1. Add unit tests for lib functions
+2. Add integration tests for commands
+3. Configure Biome and fix any issues
+4. Add npm scripts for common tasks
+
+### Phase 6: Cleanup
+
+1. Remove `tools/` directory
+2. Update any references in slash commands
+3. Final verification
+
+## Key Files to Create/Modify
+
+| File | Action |
+|------|--------|
+| `package.json` | Replace (move from tools/, update) |
+| `biome.json` | Create |
+| `tsconfig.json` | Create |
+| `vitest.config.ts` | Create |
+| `.gitignore` | Modify (add new patterns) |
+| `.tool-versions` | Modify (add pnpm) |
+| `src/index.ts` | Create |
+| `src/commands/*.ts` | Create (4 files) |
+| `src/lib/*.ts` | Create (5 files) |
+| `src/types/index.ts` | Create |
+| `tests/**/*.ts` | Create |
+| `tools/` | Delete (after migration) |
+
+## Package.json Structure
+
+```json
+{
+  "name": "honbu",
+  "version": "1.0.0",
+  "type": "module",
+  "bin": {
+    "honbu": "./src/index.ts"
+  },
+  "scripts": {
+    "dev": "node src/index.ts",
+    "test": "vitest",
+    "test:run": "vitest run",
+    "lint": "biome check .",
+    "lint:fix": "biome check --write .",
+    "format": "biome format --write .",
+    "typecheck": "tsc --noEmit"
+  },
+  "dependencies": {
+    "citty": "^0.1.6",
+    "cheerio": "^1.1.2",
+    "turndown": "^7.2.2"
+  },
+  "devDependencies": {
+    "@biomejs/biome": "^1.9.0",
+    "@types/node": "^22.0.0",
+    "@types/turndown": "^5.0.5",
+    "typescript": "^5.7.0",
+    "vitest": "^2.1.0"
+  },
+  "engines": {
+    "node": ">=23.6.0"
+  }
+}
+```
+
+## Verification
+
+1. **CLI works**: `node src/index.ts --help` shows all commands
+2. **Subcommands work**: `node src/index.ts disk overview` runs correctly
+3. **Tests pass**: `npm test` runs without errors
+4. **Lint passes**: `npm run lint` shows no errors
+5. **Types check**: `npm run typecheck` passes
+
+## Notes
+
+- Node.js 25.x runs TypeScript natively - no build step needed
+- `node-fetch` removed - use native `fetch()` (available since Node 18)
+- citty provides automatic `--help` and `--version` handling
+- Biome replaces both ESLint and Prettier with single config