Discover and explore top open-source AI tools and projects—updated daily.
tickernelzAI coding agents with persistent, searchable memory
Top 70.0% on SourcePulse
This OpenCode plugin provides AI coding agents with persistent memory using a local vector database, enabling long-term context retention across sessions. It targets OpenCode users seeking to enhance their agents' ability to recall past interactions and project details, thereby improving efficiency and continuity in development workflows.
How It Works
The system employs SQLite as the primary data store, augmented by USearch for efficient in-memory vector indexing, with an ExactScan fallback for robustness. It automatically learns user profiles, maintains a unified timeline of memories and prompts, and intelligently extracts relevant information based on user queries. The architecture supports multiple AI providers and over a dozen local embedding models, offering flexibility and privacy.
Quick Start & Requirements
"plugin": ["opencode-mem"] to ~/.config/opencode/opencode.json. The plugin downloads automatically on OpenCode startup.storagePath, embeddingModel, and AI provider details are configured in ~/.config/opencode/opencode-mem.jsonc.http://127.0.0.1:4747 for visual browsing and management.Highlighted Details
Maintenance & Community
The project actively seeks contributions for bug fixes, feature additions, and documentation improvements, with a commitment to quick pull request reviews. The codebase is described as well-structured and ready for enhancement.
Licensing & Compatibility
Licensed under the MIT License, permitting broad use, modification, and distribution, including for commercial purposes.
Limitations & Caveats
As a plugin for the OpenCode platform, its functionality is dependent on the host environment. The project is actively seeking contributions, suggesting it is still under active development and may evolve.
1 day ago
Inactive