A local semantic memory system utilizing Ollama to provide vector-based search and storage capabilities for AI agents.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install river-memory
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 river-memory 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 River Memory?
River Memory is a specialized tool designed to provide AI agents with a robust semantic search memory system. By utilizing a local Ollama instance, this Openclaw Skills entry transforms standard text into vector embeddings, allowing for nuanced information retrieval. Unlike traditional databases, River Memory focuses on meaning and context, enabling agents to find relevant information even when exact keywords are missing. This makes it a cornerstone for anyone building privacy-focused, local-first AI applications that require persistent and intelligent memory.
River Memory Use Cases
- Storing and retrieving developer snippets based on semantic context.
- Enhancing AI agent responses with project-specific long-term memory.
- Managing local knowledge bases within the Openclaw Skills framework.
- Automated cleanup and optimization of historical agent interactions.
How River Memory Works
- The system captures text input from the agent environment.
- It sends the text to a local Ollama server to generate embeddings using the nomic-embed-text model.
- The resulting vector data is stored in a local JSON memory bank.
- When queried, the system calculates semantic similarity to retrieve the most relevant entries.
- Automated routines periodically optimize the storage for peak performance.
River Memory Setup
To initialize this skill, ensure Ollama is running on your machine. Use the following commands to set up the necessary environment:
# Start the Ollama server
ollama serve
# Download the required embedding model
ollama pull nomic-embed-text
The memory store will be automatically created at ~/.openclaw/workspace/memory/vector memory.json upon first use.
River Memory Data Schema & Taxonomy
River Memory organizes its data using a structured JSON approach for local persistence. The schema is optimized for retrieval within Openclaw Skills workflows:
| Attribute | Description |
|---|---|
| Memory Content | The original text string stored in the system. |
| Vector Embedding | The high-dimensional array generated by Ollama. |
| Storage Location | Resides in the local user directory for privacy. |
| Metadata | Includes timestamps and management flags for cleanup. |
River Memory Advanced Features
- Local-only processing for maximum data privacy.
- Support for the nomic-embed-text model via Ollama.
- Automatic memory pool optimization and scheduled cleanup.
- Deep integration with other Openclaw Skills for multi-agent coordination.
SKILL.md
Loading
Related Openclaw Skills
Image Generator Skill
A local image generation workflow supporting Stable Diffusion 1.5 and SDXL models integrated with non-blocking subagent execution.
Exa Search & Research MCP
A powerful neural search integration that enables AI agents to perform advanced web crawling and deep research using the Exa protocol.
OpenProse AI Virtual Machine
OpenProse is a specialized programming language that transforms LLMs into high-fidelity virtual machines capable of orchestrating complex multi-agent sessions.
Tunneling (TinyFish)
A seamless SSH-based tunneling skill to create public HTTPS URLs for local applications without any signup or software installation.
Gradient Inference for DigitalOcean
Integrate DigitalOcean's Gradient Serverless Inference into your AI agent workflows for high-performance LLM chat and image generation.
Gradient Knowledge Base & RAG
A community-driven skill for building Retrieval-Augmented Generation (RAG) pipelines using DigitalOcean Gradient and Spaces storage.
Featured*
