🌟 Community Call to Action: Have you made improvements or additions to this system? Submit a pull request! Every contributor will be properly credited in the final product.
A persistent local memory for AI, LLMs, or Copilot in VS Code.
Full Feature Parity Release -- PAM is now fully in sync with the internal
production system. Every feature from months of real-world use is now public.
Core Identity System -- Distills a complete profile of who the user is
across all conversations. Pulls from curated memories, the OpenWebUI memory
table, and archived databases. Updates incrementally rather than starting
from scratch each time.
Automated Memory Maintenance -- Background tasks reformat old memories,
detect contradictions and updates, and link orphaned memories to their source
conversations. Includes a backlog processor that deduplicates entries, fills
in missing metadata, and re-ranks memory importance using the LLM.
Task Coordinator -- Centralized scheduler replaces ad-hoc sleep loops.
Proper concurrency control with per-database locking and LLM call gating.
Detects user activity and waits until idle to run heavy work.
Vision Support -- Precomputes image embeddings via a separate vision
server so the main LLM does not need mmproj loaded. Caches embeddings for
cross-turn injection. Works with llama.cpp, LM Studio, and any
OpenAI-compatible embedding provider.
Cross-System Deduplication -- When promoting short-term memories to
long-term storage, checks for exact and semantic duplicates. Merges related
memories instead of creating redundant copies.
Heartbeat-Based Idle Detection -- The system respects when you are
actively using it and delays heavy background tasks until you are idle.
New Utility Tools -- Interactive model-specific memory deletion
(pam_delete_model_memories.py), memory export to text files
(export_memories.py), and MCP tool call filtering (filter_tool_calls.py).
Normalization Migration -- One-time migration that normalizes user and
model IDs for consistent multi-user isolation.
Upgrade from v1.5.0: Drop-in replacement. Replace your existing PAM files
with the new versions. The database schema is backward compatible.
What's New in v1.5.0 (March 28, 2026)
Major Architectural Rewrite: OpenWebUI-Native Integration
✅ OpenWebUI-first design - AI Memory System now deeply integrated into OpenWebUI via plugin (primary deployment method)
✅ Advanced short-term memory - sophisticated memory extraction, filtering, and injection for chat conversations
✅ User ID & Model ID isolation - strict multi-tenant support with configurable enforcement for security and tracking
✅ Complete system portability - all hardcoded paths replaced with environment variables (works anywhere)
✅ Generic class names - removed all Friday-specific branding (FridayMemorySystem → AIMemorySystem)
✅ Production-ready - enhanced error handling, validation, and logging throughout
Upgrade from v1.1.0: See CHANGELOG.md for migration guide.
# Linux/macOS
pip install git+https://github.com/savantskie/persistent-ai-memory.git
# Windows (same command, just use Command Prompt or PowerShell)
pip install git+https://github.com/savantskie/persistent-ai-memory.git
First Validation
python tests/test_health_check.py
Expected output:
[✓] Imported ai_memory_core
[✓] Found embedding_config.json
[✓] System health check passed
[✓] All health checks passed! System is ready to use.
💡 What This System Does
Persistent AI Memory provides sophisticated memory management for AI assistants:
📝 OpenWebUI Short-Term Memory Plugin - Intelligent memory extraction and injection directly in chat conversations
require_user_id/require_model_id: true → Strict mode (recommended for production, security-focused, or multi-user systems)
require_user_id/require_model_id: false → Use defaults instead (simpler for single-user/single-model setups)
For AI Assistants: Auto-Fill in System Prompt
To make your AI automatically provide these values, add this to its system prompt:
When using memory system tools (store_memory, search_memories, etc.),
ALWAYS include these parameters:
- user_id='your_user_identifier' (e.g., 'nate_user_1')
- model_id='your_model_name' (e.g., 'llama-2:7b' or 'gpt-4')
If the actual values are unknown, use safe defaults:
- user_id='default_user'
- model_id='default_model'
This isolates memories per user and tracks which AI model generated each memory.
Examples
With user_id and model_id:
# Memories are stored with full isolation
await system.store_memory(
"User likes Python",
user_id="alice",
model_id="gpt-4"
)
# Search returns only this user's memories for this model
results = await system.search_memories(
"programming",
user_id="alice",
model_id="gpt-4"
)
Primary deployment method - Deep integration for sophisticated memory management:
Deploy ai_memory_short_term.py as an OpenWebUI Function
Automatically extracts memories from conversations
Intelligently injects relevant memories before AI response
Configurable memory scoring, filtering, and injection preferences
No additional setup required beyond copying file into OpenWebUI Functions editor
Installation:
In OpenWebUI: Settings → Functions → +New Function
Paste entire ai_memory_short_term.py file
Set trigger to Inlet (runs before model response)
Configure memory preferences via function settings
2. MCP Server (Alternative Platforms)
Use with any MCP-compatible AI assistant (Claude, custom integrations, etc.):
# Via mcpo
python -m ai_memory_mcp_server
# Or make streamable for OpenWebUI's alternative integration
# (OpenWebUI supports both plugin and streamable MCP methods)
3. Standalone Library (Custom Implementations)
Use memory capabilities directly in your Python code:
from ai_memory_core import AIMemorySystem
system = AIMemorySystem()
await system.store_memory("Important information", user_id="user1", model_id="model1")
results = await system.search_memories("query", user_id="user1", model_id="model1")
🛠️ Development & Examples
Ready-to-use examples:
python examples/basic_usage.py # Store and search memories
python examples/advanced_usage.py # Conversation tracking and tool logging
python examples/performance_tests.py # Benchmark operations
अभी टूल सूची नहीं है। रिमोट सर्वर को पहले हेल्थ चेक के बाद सूची मिलती है।
FAQ
सवाल
Persistent Ai Memory कैसे इंस्टॉल करें?
इस पेज पर Cursor, Claude Desktop, Claude Code या VS Code के लिए stdio कॉन्फ़िग कॉपी करें। यह पब्लिशर का इंस्टॉल इस्तेमाल करता है, PlugMCP प्रॉक्सी नहीं।
क्या Persistent Ai Memory के लिए API key चाहिए?
इस रिकॉर्ड में कोई ज़रूरी एनवायरनमेंट वेरिएबल दर्ज नहीं है।
Persistent Ai Memory कौन सा ट्रांसपोर्ट इस्तेमाल करता है?
Persistent Ai Memory stdio पर लोकल चलता है।
Persistent Ai Memory की पिछली जाँच कब हुई थी?
पिछली जाँच 2026-10-26 को हुई। A persistent local memory for AI, LLMs, or Copilot in VS Code.