A local semantic memory skill that enables AI agents to store, search, and recall conversation context using vector embeddings without external API keys.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install openclaw-memory-qdrant
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install openclaw-memory-qdrant using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is memory-qdrant?
The memory-qdrant skill is a privacy-focused local memory solution designed to give AI agents long-term recall capabilities. By utilizing the Qdrant vector database and Transformers.js for local embeddings, this skill allows your agent to perform semantic searches over past interactions, ensuring that context is never lost across sessions. It is a vital addition to the ecosystem of Openclaw Skills for developers who prioritize data sovereignty and offline performance.
This skill operates entirely on your local machine, downloading a lightweight embedding model on its first run. It bridges the gap between stateless LLM interactions and personalized AI assistants by providing a structured way to manage preferences, facts, and historical data. Whether you are building a complex coding assistant or a personal research tool, integrating this into your Openclaw Skills workflow ensures a more intelligent and context-aware experience.
memory-qdrant Use Cases
- Maintaining user preferences and architectural decisions across long-term development projects.
- Retrieving relevant code snippets or documentation from previous conversations.
- Automatically capturing and organizing facts during research-heavy tasks.
- Reducing token overhead by searching for specific context rather than feeding entire histories into the prompt.
How memory-qdrant Works
- Upon activation, the skill initializes a local Qdrant instance and loads the Xenova/all-MiniLM-L6-v2 embedding model.
- When data is saved through the memory_store tool, the text is converted into a vector embedding and stored in the database.
- During a query, the memory_search tool generates a vector for the search term and identifies the most semantically similar memories.
- If auto-recall is enabled, the system automatically fetches and injects relevant historical context into the current conversation.
- The memory_forget tool allows for precise maintenance by removing specific memories via ID or text query.
memory-qdrant Setup
To add this to your collection of Openclaw Skills, run the following command:
clawhub install memory-qdrant
Then, enable the plugin in your OpenClaw configuration file:
{
"plugins": {
"memory-qdrant": {
"enabled": true,
"persistToDisk": true
}
}
}
memory-qdrant Data Schema & Taxonomy
The skill organizes data locally to ensure privacy and speed. Below is the structure of how information is managed:
| Component | Description |
|---|---|
| Storage Path | Defaults to ~/.openclaw-memory/ for persistent data. |
| Vector Model | Xenova/all-MiniLM-L6-v2 (~25MB model). |
| Metadata | Includes original text, unique memory IDs, and optional categories. |
| Database | Qdrant (supports in-memory, disk-persistent, or external server modes). |
memory-qdrant Advanced Features
- PII Protection: Automatically filters out sensitive information like emails and phone numbers during capture.
- Auto-Capture: An opt-in feature that intelligently records facts and decisions based on semantic triggers.
- Multi-Mode Storage: Choose between volatile in-memory sessions or persistent disk storage for long-term retention.
- External Integration: Support for connecting to an external Qdrant server for advanced infrastructure needs.
SKILL.md
Loading
Related Openclaw Skills
memory-qdrant
A local semantic memory plugin for Openclaw Skills providing vector-based context storage and recall without external API dependencies.
Feishu Edge TTS Voice Sender (Windows)
A specialized Windows skill for converting text to speech and sending it as native Feishu voice bubbles.
AgentConstitution Skill
A decentralized governance framework on Base Sepolia that allows AI agents to follow rules, log actions, and prove compliance on-chain.
GemBox Skill
A comprehensive developer tool for building .NET applications that process Excel, Word, PDF, and email files using GemBox components.
Semantic Memory
A locally-hosted semantic memory plugin that allows OpenClaw agents to store and recall conversation context using vector search and Transformers.js.
Vault
A secure local password manager that uses AES-256-GCM encryption to protect API keys and credentials via a command-line interface.
Featured*
