opencode-mem

name	opencode-mem
description	Persistent memory system with vector database for long-term context retention across sessions. Features dual memory scopes, web interface, auto-capture system, and multi-provider AI support.
license	MIT
compatibility	opencode
metadata	{"source":"https://github.com/tickernelz/opencode-mem","adapted-for":"opencode"}

Opencode Mem

Persistent memory system with vector database for long-term context retention across sessions.

Overview

A persistent memory system for AI coding agents that enables long-term context retention across sessions using local vector database technology. Features dual memory scopes, web interface, auto-capture system, and multi-provider AI support.

When to Use

Use this skill when:

• You need long-term memory persistence across sessions
• Working on complex projects that require context retention
• Want to automatically capture important information
• Need to search through historical context and decisions
• Working with multiple providers and want unified memory

Core Features

Dual Memory Scopes

1. Global Memory: Shared across all projects and sessions
2. Project Memory: Specific to the current repository

Vector Database

• Local storage using vector embeddings
• Semantic search capabilities
• Automatic indexing of content
• Efficient retrieval of relevant information

Web Interface

• Browser-based UI for memory management
• Search and browse capabilities
• Memory editing and organization
• Visualization of memory relationships

Auto-Capture System

• Automatic capture of important context
• Configurable triggers and rules
• Smart filtering of relevant information
• Integration with tool outputs

Multi-Provider Support

• Works with multiple AI providers
• Unified memory interface
• Provider-specific optimizations
• Cross-provider context sharing

Workflow

Memory Creation

1. Automatic Capture

   • Tool outputs are automatically indexed
   • Important decisions are captured
   • User interactions are stored when relevant

2. Manual Storage

   /mem store "Important project decision" --tags=architecture,decision
   

3. File Integration

   /mem import @README.md --type=documentation
   

Memory Retrieval

1. Semantic Search

   /mem search "database connection issues"
   

2. Tag-based Search

   /mem find --tag=architecture
   

3. Recent Memory

   /mem recent --limit=5
   

Memory Management

1. Organization

   /mem tag "memory-id" architecture,backend
   

2. Updates

   /mem update "memory-id" "Updated content"
   

3. Deletion

   /mem delete "memory-id"
   

Best Practices

Content Strategy

• Store high-value information that's hard to recreate
• Focus on decisions, rationale, and complex context
• Avoid storing transient or easily reproducible data
• Use tags for better organization and retrieval

Tagging System

Recommended tags:

• decision: Important architectural or design decisions
• error: Error patterns and solutions
• pattern: Reusable patterns and practices
• requirement: User requirements and constraints
• research: Research findings and insights

Search Optimization

• Use specific, descriptive search terms
• Combine keywords with tags when possible
• Review search results to improve future queries
• Update memory content to improve searchability

Integration Points

With Other Skills

• agent-memory: Complements file-based memory with vector search
• napkin: Stores high-level patterns and recurring issues
• systematic-debugging: Captures debugging insights and solutions

Tool Integration

• Automatic indexing of tool outputs
• Context-aware memory suggestions
• Integration with file operations
• Shell command result storage

Configuration

Memory Settings

/mem config --scope=project --auto-capture=true

Provider Setup

/mem provider openai --api-key=your-key

Web Interface

/mem web --port=3000

Example Usage

Project Onboarding

/mem import @docs/architecture.md --type=architecture
/mem import @README.md --type=documentation
/mem search "project setup"

Debugging Session

/mem store "Fixed memory leak by implementing proper cleanup" --tags=debug,error
/mem search "memory leak" --related=true

Knowledge Building

/mem store "User requested real-time notifications feature" --tags=requirement,user
/mem find --tag=pattern --limit=10

Web Interface Features

Dashboard

• Overview of memory statistics
• Recent memories and updates
• Search quick access
• Configuration controls

Search Interface

• Full-text and semantic search
• Filter by tags and types
• Sort by relevance, date, or usage
• Preview and quick actions

Memory Management

• Create, edit, and delete memories
• Tag management and organization
• Bulk operations and exports
• Import from various sources

Performance Considerations

Indexing

• Automatic background indexing
• Incremental updates for efficiency
• Configurable indexing frequency
• Manual reindex when needed

Search Optimization

• Cached search results for common queries
• Efficient vector similarity calculations
• Pagination for large result sets
• Search result relevance tuning

Troubleshooting

Memory Not Found

• Check if auto-capture is enabled
• Verify content was properly stored
• Try broader search terms
• Check both global and project scopes

Performance Issues

• Review indexing settings
• Check vector database size
• Optimize memory content
• Consider scope separation

Web Interface Problems

• Verify port availability
• Check network connectivity
• Review browser compatibility
• Clear cache and reload