hybrid-memory
Hybrid Memory Skill
A self-contained hybrid memory system inspired by ZeroClaw's architecture. Combines vector embeddings (semantic search) with FTS5 keyword search (BM25 scoring) for powerful memory recall.
Purpose
Give agents persistent memory with:
- Semantic search - Find memories by meaning, not just keywords
- Keyword search - Traditional BM25 text search via SQLite FTS5
- Hybrid scoring - Weighted combination of vector and keyword similarity
- Zero external dependencies - Pure SQLite + local embeddings
Architecture
┌─────────────────────────────────────────────────────────────┐
│ Hybrid Memory │
├─────────────────────────────────────────────────────────────┤
│ ┌──────────────┐ ┌──────────────┐ ┌──────────────────┐ │
More from winsorllc/upgraded-carnival
vector-memory
Vector-based semantic memory using embeddings for intelligent recall. Store and search memories by meaning rather than keywords. Use when you need semantic search, similar document retrieval, or context-aware memory.
136model-router
Route requests between different LLM providers and models. Configure routing rules, fallback providers, and model-specific parameters inspired by ZeroClaw and OpenClaw model routing systems.
70rss-monitor
Monitor RSS/Atom feeds and blogs for new content using feedparser.
63schedule-task
Create and manage scheduled shell tasks. Use when: automating recurring operations. NOT for: sending messages (use cron agent).
58rss-reader
Read and parse RSS/Atom feeds. Use when: user wants to subscribe to feeds, get latest articles, or monitor news sources.
57video-frames
Production-grade video frame extraction with thumbnail grids, GIF creation, and batch frame processing. Includes intelligent quality presets, progress tracking, and comprehensive error handling.
40