superlocalmemory 3.4.16 → 3.4.18

package/docs/v2-archive/PATTERN-LEARNING.md

@@ -1,228 +0,0 @@
-# Pattern Learning
-
-## Overview
-
-The Pattern Learner is Layer 4 of the SuperLocalMemory system. It automatically learns user preferences, coding style, and terminology from stored memories — all processed locally with no external API calls.
-
-**No external APIs. All processing is local.**
-
-> For technical details on the learning algorithms, see our published research: https://zenodo.org/records/18709670
-
----
-
-## What It Learns
-
-Pattern learning analyzes your saved memories and builds a profile of your preferences across multiple dimensions:
-
-- **Technology preferences** — Which frameworks, languages, and tools appear most often in your memories
-- **Architecture choices** — Your preferred patterns (e.g., REST vs GraphQL, microservices vs monolith)
-- **Security approaches** — Your go-to authentication and authorization patterns
-- **Coding style** — Performance vs readability, testing approach, code organization
-- **Domain terminology** — How you define common terms in your project context
-
-Each detected preference gets a confidence score based on the supporting evidence. The system is conservative by design — low confidence with small datasets is correct behavior, not a bug.
-
----
-
-## Usage
-
-### Command Line Interface
-
-```bash
-# Full pattern update (run weekly or after bulk imports)
-python pattern_learner.py update
-
-# List learned patterns
-python pattern_learner.py list [min_confidence]
-# Example: python pattern_learner.py list 0.7
-
-# Get context formatted for AI assistant injection
-python pattern_learner.py context [min_confidence]
-# Example: python pattern_learner.py context 0.6
-
-# Show statistics
-python pattern_learner.py stats
-```
-
-### Python API
-
-```python
-from pattern_learner import PatternLearner
-
-learner = PatternLearner()
-
-# Full pattern analysis
-counts = learner.weekly_pattern_update()
-# Returns: {'preferences': 5, 'styles': 3, 'terminology': 4}
-
-# Incremental update for new memory
-learner.on_new_memory(memory_id=42)
-
-# Query patterns
-patterns = learner.get_patterns(min_confidence=0.7)
-
-# Get formatted context for AI injection
-context = learner.get_identity_context(min_confidence=0.7)
-```
-
----
-
-## Example Output
-
-### Pattern List
-
-```
-Type            Category     Pattern                        Confidence   Evidence
------------------------------------------------------------------------------------------------
-style           general      Optimization Priority: Perfo...    90%        15
-preference      frontend     Frontend Framework: Next.js...    85%        12
-preference      backend      Database: PostgreSQL              72%        8
-terminology     general      Optimize: Performance optimi...    90%        15
-style           general      Error Handling: Explicit err...    78%        9
-```
-
-### AI Context Injection
-
-```markdown
-## Working with User - Learned Patterns
-
-**Technology Preferences:**
-- **Frontend Framework:** Next.js over React (confidence: 85%, 12 examples)
-- **Database:** PostgreSQL (confidence: 72%, 8 examples)
-
-**Coding Style:**
-- **Optimization Priority:** Performance over readability (confidence: 90%, 15 examples)
-- **Error Handling:** Explicit error boundaries (confidence: 78%, 9 examples)
-
-**Terminology:**
-- **Optimize:** Performance optimization (speed/latency) (confidence: 90%, 15 examples)
-```
-
----
-
-## Confidence Calibration
-
-| Dataset Size | Expected Confidence | Recommended Threshold |
-|-------------|--------------------|-----------------------|
-| Under 20 memories | 10-40% | Use 0.1 for exploration |
-| 20-100 memories | 40-70% | Use 0.5 for reliable patterns |
-| Over 100 memories | 70-95% | Use 0.7 for high-confidence patterns |
-
-The system is designed to be conservative. Low confidence with small datasets is correct behavior.
-
----
-
-## Pattern Update Schedule
-
-### Incremental (on `/remember`)
-- Quick check for pattern updates on every new memory
-- For large datasets, defers to weekly consolidation
-
-### Weekly Consolidation
-- Full re-analysis of all memories
-- Recommended cron job:
-  ```bash
-  # Every Sunday at 2 AM
-  0 2 * * 0 cd ~/.superlocalmemory && python3 pattern_learner.py update >> pattern_updates.log 2>&1
-  ```
-
----
-
-## Integration with AI Assistants
-
-### Automatic Context Injection
-
-The `get_identity_context()` method formats patterns for any AI assistant's context window:
-
-```python
-learner = PatternLearner()
-identity_context = learner.get_identity_context(min_confidence=0.7)
-
-# Inject into system prompt or context (works with Claude, GPT, or any assistant)
-prompt = f"""
-{identity_context}
-
-[Rest of your prompt...]
-"""
-```
-
-**Works with:** Claude CLI, GPT, local LLMs, or any AI assistant that accepts context.
-
-### Benefits
-1. **Personalized responses:** AI understands your technology preferences
-2. **Consistent terminology:** AI uses terms the way you define them
-3. **Style matching:** AI respects your coding style priorities
-4. **Efficient communication:** No need to repeat preferences every session
-
----
-
-## Privacy & Security
-
-- **No external API calls** — all processing is local
-- **All data stays local** — stored in `~/.superlocalmemory/`
-- **SQL injection protected** — parameterized queries throughout
-- **No credentials stored**
-- **No network access required**
-
----
-
-## Limitations
-
-1. **English-optimized:** Pattern detection works best with English text
-2. **Technical domains:** Works best with technical and coding memories
-3. **Minimum samples:** Requires a few mentions before a pattern is detected
-4. **Conservative scoring:** Low confidence with small datasets is by design
-
----
-
-## Troubleshooting
-
-### "No patterns learned yet"
-- **Cause:** Not enough memories or technical content
-- **Solution:** Add 10-20 memories with technical details about your preferences
-
-### Low confidence scores
-- **Cause:** Small dataset or mixed evidence
-- **Solution:** Normal behavior. Lower the threshold (0.1-0.3) or add more memories
-
-### Wrong preference detected
-- **Cause:** More mentions of alternative technology than your true preference
-- **Solution:** Add more memories explicitly stating your correct preference
-
-### Pattern not updating
-- **Cause:** Weekly update not run
-- **Solution:** Run `python pattern_learner.py update` manually
-
----
-
-## Testing
-
-Sample memories for verifying pattern detection:
-
-```bash
-# Add sample memories
-python memory_store.py add "Built with Next.js for better performance" --tags nextjs
-python memory_store.py add "Using Next.js on all new projects" --tags nextjs
-python memory_store.py add "Optimized API from slow to fast" --tags performance
-python memory_store.py add "Performance is critical for UX" --tags performance
-python memory_store.py add "When I say optimize, I mean make it faster" --tags terminology
-
-# Run pattern learning
-python pattern_learner.py update
-
-# Check results
-python pattern_learner.py list 0.1
-python pattern_learner.py context 0.1
-```
-
----
-
-## License
-
-Part of SuperLocalMemory system. See main project LICENSE.
-
----
-
-**Implementation Status:** Complete and tested
-
-**Last Updated:** 2026-02-05

package/docs/v2-archive/PROFILES-GUIDE.md

@@ -1,453 +0,0 @@
-# SuperLocalMemory V2 - Profile Management Guide
-
-## Overview
-
-**Profiles** let you maintain completely separate memory systems for different contexts or personalities.
-
-**Use cases:**
-- **Work vs Personal**: Keep professional and personal memories separate
-- **Client-specific**: Isolated memories for each client project
-- **Experimentation**: Test profile for trying new features
-- **Personality switching**: Different "AI Master" personas with unique learned patterns
-
----
-
-## Quick Start
-
-### List Profiles
-```bash
-python ~/.superlocalmemory/memory-profiles.py list
-```
-
-### Show Current Profile
-```bash
-python ~/.superlocalmemory/memory-profiles.py current
-```
-
-### Create New Profile
-```bash
-# Empty profile
-python ~/.superlocalmemory/memory-profiles.py create work \
-  --description "Work projects"
-
-# Copy current memories to new profile
-python ~/.superlocalmemory/memory-profiles.py create personal \
-  --from-current
-```
-
-### Switch Profile
-```bash
-python ~/.superlocalmemory/memory-profiles.py switch work
-
-# ⚠️ IMPORTANT: Restart Claude CLI after switching!
-```
-
-### Delete Profile
-```bash
-python ~/.superlocalmemory/memory-profiles.py delete old-profile
-```
-
----
-
-## Profile Isolation
-
-Each profile has its own:
-- ✅ **Database** (memory.db) - Completely separate memories
-- ✅ **Graph data** - Independent knowledge graph
-- ✅ **Learned patterns** - Different identity profiles
-- ✅ **Compressed archives** - Separate storage
-- ✅ **Configuration** - Profile-specific settings
-
-**Profiles DO NOT share:**
-- Memories
-- Graphs
-- Patterns
-- Any data
-
----
-
-## Use Cases
-
-### 1. Work vs Personal
-
-```bash
-# Create work profile
-python ~/.superlocalmemory/memory-profiles.py create work \
-  --description "Professional coding projects"
-
-# Create personal profile from current
-python ~/.superlocalmemory/memory-profiles.py create personal \
-  --from-current --description "Personal learning and experiments"
-
-# Switch based on context
-python ~/.superlocalmemory/memory-profiles.py switch work   # During work hours
-python ~/.superlocalmemory/memory-profiles.py switch personal  # Personal time
-```
-
-**Result:**
-- Work profile learns: Enterprise patterns, client tech stacks, professional terminology
-- Personal profile learns: Hobby projects, experimental tech, casual coding style
-
----
-
-### 2. Client-Specific Profiles
-
-```bash
-# Create profile per client
-python ~/.superlocalmemory/memory-profiles.py create client-acme \
-  --description "Acme Corp project"
-
-python ~/.superlocalmemory/memory-profiles.py create client-contoso \
-  --description "Contoso Inc project"
-
-# Switch when working on different clients
-python ~/.superlocalmemory/memory-profiles.py switch client-acme
-```
-
-**Benefits:**
-- Client data never mixes
-- Each client gets learned patterns specific to their codebase
-- Easy to archive completed projects
-
----
-
-### 3. Learning & Experimentation
-
-```bash
-# Create test profile
-python ~/.superlocalmemory/memory-profiles.py create experimental \
-  --description "Testing new features"
-
-# Switch to test profile
-python ~/.superlocalmemory/memory-profiles.py switch experimental
-
-# Try new things without polluting main profile
-# ...test features...
-
-# Switch back when done
-python ~/.superlocalmemory/memory-profiles.py switch default
-```
-
----
-
-### 4. Different AI Personalities
-
-```bash
-# Strict architect profile
-python ~/.superlocalmemory/memory-profiles.py create architect \
-  --description "Rigorous architecture-first approach"
-
-# Rapid prototyper profile
-python ~/.superlocalmemory/memory-profiles.py create prototyper \
-  --description "Fast iteration, MVP-first approach"
-
-# Use based on project phase
-python ~/.superlocalmemory/memory-profiles.py switch architect  # Planning
-python ~/.superlocalmemory/memory-profiles.py switch prototyper  # MVP development
-```
-
-**Result:** Each profile learns different patterns:
-- Architect: Detailed docs, extensive planning, clean architecture
-- Prototyper: Quick iterations, pragmatic choices, speed over perfection
-
----
-
-## Profile Operations
-
-### Create Empty Profile
-
-Creates fresh V2 database with no memories:
-
-```bash
-python ~/.superlocalmemory/memory-profiles.py create new-profile
-```
-
-**When to use:**
-- Starting completely fresh context
-- Client work (no prior context)
-- Experimentation (clean slate)
-
----
-
-### Create from Current
-
-Copies current memory system to new profile:
-
-```bash
-python ~/.superlocalmemory/memory-profiles.py create backup --from-current
-```
-
-**When to use:**
-- Branching current memories
-- Creating backup before major changes
-- Testing with real data
-
----
-
-### Switch Profile
-
-Changes active profile (requires Claude CLI restart):
-
-```bash
-python ~/.superlocalmemory/memory-profiles.py switch work
-
-# Output:
-# Switching from 'default' to 'work'...
-#   Saving current state to profile 'default'...
-#     ✓ Saved to ~/.superlocalmemory/profiles/default
-#   Loading profile 'work'...
-#     ✓ Loaded from ~/.superlocalmemory/profiles/work
-# ✅ Switched to profile: work
-#
-# ⚠️  IMPORTANT: Restart Claude CLI to use new profile!
-```
-
-**⚠️ Must restart Claude CLI after switching!**
-
----
-
-### Delete Profile
-
-Permanently removes a profile:
-
-```bash
-python ~/.superlocalmemory/memory-profiles.py delete old-profile
-
-# Prompts: Type profile name 'old-profile' to confirm:
-```
-
-**Safety:**
-- Cannot delete 'default' profile
-- Cannot delete active profile
-- Requires confirmation (type profile name)
-- Use `--force` to skip confirmation
-
----
-
-### Rename Profile
-
-```bash
-python ~/.superlocalmemory/memory-profiles.py rename old-name new-name
-```
-
-**Cannot rename:**
-- 'default' profile
-- Active profile (switch first)
-
----
-
-## Profile Storage
-
-Profiles stored in:
-```
-~/.superlocalmemory/profiles/
-├── work/
-│   ├── memory.db
-│   ├── config.json
-│   └── vectors/
-├── personal/
-│   ├── memory.db
-│   ├── config.json
-│   └── vectors/
-└── client-acme/
-    ├── memory.db
-    ├── config.json
-    └── vectors/
-```
-
-**Active profile** stored in main location:
-```
-~/.superlocalmemory/
-├── memory.db          ← Currently active profile
-├── config.json
-└── vectors/
-```
-
----
-
-## Profile Configuration
-
-Stored in `~/.superlocalmemory/profiles.json`:
-
-```json
-{
-  "profiles": {
-    "default": {
-      "name": "default",
-      "description": "Default memory profile",
-      "created_at": "2026-02-05T10:00:00",
-      "last_used": "2026-02-05T14:00:00",
-      "created_from": "empty"
-    },
-    "work": {
-      "name": "work",
-      "description": "Work projects",
-      "created_at": "2026-02-05T11:00:00",
-      "last_used": "2026-02-05T13:00:00",
-      "created_from": "empty"
-    }
-  },
-  "active_profile": "work"
-}
-```
-
----
-
-## Best Practices
-
-### 1. Use Descriptive Names
-```bash
-# Good
-python ~/.superlocalmemory/memory-profiles.py create client-acme-api
-python ~/.superlocalmemory/memory-profiles.py create personal-learning
-
-# Avoid
-python ~/.superlocalmemory/memory-profiles.py create profile1
-python ~/.superlocalmemory/memory-profiles.py create temp
-```
-
-### 2. Document Profile Purpose
-```bash
-python ~/.superlocalmemory/memory-profiles.py create project-x \
-  --description "Confidential client X - Next.js + PostgreSQL"
-```
-
-### 3. Switch Profiles at Context Boundaries
-```bash
-# Start of work day
-python ~/.superlocalmemory/memory-profiles.py switch work
-
-# End of work day
-python ~/.superlocalmemory/memory-profiles.py switch personal
-```
-
-### 4. Backup Before Major Changes
-```bash
-# Create backup profile
-python ~/.superlocalmemory/memory-profiles.py create backup-$(date +%Y%m%d) \
-  --from-current
-
-# Make changes in main profile
-# ...
-
-# If needed, switch to backup
-python ~/.superlocalmemory/memory-profiles.py switch backup-20260205
-```
-
-### 5. Archive Completed Projects
-```bash
-# Rename completed project
-python ~/.superlocalmemory/memory-profiles.py rename client-acme \
-  archived-client-acme-202602
-
-# Create fresh profile for new project
-python ~/.superlocalmemory/memory-profiles.py create client-newco
-```
-
----
-
-## Integration with Reset Commands
-
-**Profiles + Reset = Powerful Combination**
-
-### Reset Current Profile Only
-```bash
-# Soft reset active profile
-python ~/.superlocalmemory/memory-reset.py soft
-
-# Active profile cleared, others untouched
-```
-
-### Reset Specific Profile
-```bash
-# Switch to profile
-python ~/.superlocalmemory/memory-profiles.py switch old-project
-
-# Reset it
-python ~/.superlocalmemory/memory-reset.py soft
-
-# Switch back
-python ~/.superlocalmemory/memory-profiles.py switch default
-```
-
-### Create Clean Profile
-```bash
-# Create empty profile
-python ~/.superlocalmemory/memory-profiles.py create clean-slate
-
-# Switch to it
-python ~/.superlocalmemory/memory-profiles.py switch clean-slate
-
-# Now work in completely clean environment
-```
-
----
-
-## Troubleshooting
-
-### "Profile not found"
-Check available profiles:
-```bash
-python ~/.superlocalmemory/memory-profiles.py list
-```
-
-### "Changes not reflecting"
-Restart Claude CLI after switching profiles:
-```bash
-# Exit Claude CLI
-# Restart Claude CLI
-```
-
-### "Cannot delete active profile"
-Switch to different profile first:
-```bash
-python ~/.superlocalmemory/memory-profiles.py switch default
-python ~/.superlocalmemory/memory-profiles.py delete unwanted
-```
-
-### "Profile directory not found"
-Profile will be created on first switch:
-```bash
-python ~/.superlocalmemory/memory-profiles.py switch new-profile
-# Creates directory automatically
-```
-
----
-
-## FAQ
-
-**Q: Do profiles share learned patterns?**
-A: No. Each profile learns independently.
-
-**Q: Can I merge profiles?**
-A: Not directly. You can manually copy memories between profiles.
-
-**Q: What happens to cron jobs with multiple profiles?**
-A: Cron jobs run on currently active profile. Switch profile to run jobs on it.
-
-**Q: Can I have different compression settings per profile?**
-A: Yes. Each profile has its own `config.json`.
-
-**Q: How much space do profiles use?**
-A: ~5MB per profile with 100 memories (before compression).
-
-**Q: Can I backup all profiles at once?**
-A: Yes, backup `~/.superlocalmemory/profiles/` directory.
-
----
-
-## Quick Reference
-
-| Command | Purpose |
-|---------|---------|
-| `list` | Show all profiles |
-| `current` | Show active profile |
-| `create <name>` | Create empty profile |
-| `create <name> --from-current` | Copy current to new profile |
-| `switch <name>` | Change active profile (requires restart) |
-| `delete <name>` | Remove profile (requires confirmation) |
-| `rename <old> <new>` | Rename profile |
-
----
-
-**Remember:** Always restart Claude CLI after switching profiles!