vector-memory
Smart memory search with automatic vector fallback. Uses semantic embeddings when available, falls back to built-in search otherwise. Zero configuration - works immediately after ClawHub install. No setup required - just install and memory_search works immediately, gets better after optional sync.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/bluepointdigital/vector-memoryWhat This Skill Does
Vector Memory is an intelligent search engine designed for the OpenClaw ecosystem, providing a seamless bridge between traditional keyword-based searching and advanced semantic vector retrieval. By utilizing an automatic selection strategy, the skill ensures that users get the best possible search results regardless of their configuration state. When you first install the skill, it functions as a high-speed keyword search utility. Once you perform a synchronization operation, it automatically upgrades to leverage neural embeddings, allowing the agent to understand concepts, synonyms, and the underlying intent of your queries rather than relying solely on exact word matches.
Installation
The process is streamlined for immediate use within the OpenClaw environment. You can install the skill directly via the ClawHub command line interface. Simply run:
npx clawhub install vector-memory
Once executed, the skill is immediately active. No complex configuration files are required, as the system defaults to its built-in search mode upon initial installation. For those requiring enhanced semantic capabilities, running node vector-memory/smart_memory.js --sync will index your local memory, enabling the vector search engine to begin processing and matching based on semantic intent.
Use Cases
This skill is ideal for knowledge management, project documentation retrieval, and personal knowledge base interaction. Use it to:
- Retrieve obscure concepts from project notes when you cannot remember specific keywords.
- Automatically index and search through massive markdown-based documentation folders.
- Maintain a consistent search interface that scales from a few local files to thousands of content chunks.
- Bridge the gap between static technical notes and an active, AI-assisted assistant.
Example Prompts
- "Search my notes for principles regarding autonomous system design; look beyond just the literal word 'autonomous'."
- "Memory search for the last discussion we had about the project architecture update, specifically regarding the database migration path."
- "Retrieve everything I've documented regarding authentication protocols for our API integration."
Tips & Limitations
To get the most out of Vector Memory, ensure you run the sync command whenever you make significant changes to your memory directory. While the built-in search is incredibly fast and reliable for simple keyword lookups, the vector-based approach is superior for long-form conceptual queries. If your dataset grows beyond 1,000 chunks, consider moving from the default local storage to an external pgvector instance to maintain optimal latency. Always keep your MEMORY_DIR correctly pointed if you are working with non-standard file structures.
Metadata
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skillPaste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-bluepointdigital-vector-memory": {
"enabled": true,
"auto_update": true
}
}
}Tags(AI)
Flags: file-read, file-write