OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Agent Memory & RAG

Kektordb

Kektordb is a self-hostable AI-agent memory system that combines HNSW vector search with a temporal knowledge graph and cognitive features for storing, retrieving and evolving memories.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Kektordb — OpenViST project profile in Agent Memory & RAG.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Kektordb when an agent needs persistent, searchable memory and its development stack can use the documented MCP server, Python client, REST API or embedded Go engine.

Who it is for

  • AI-agent developers who need context to persist across sessions.
  • Go developers embedding a memory engine in an application.
  • Python developers integrating memory through the documented client and REST API.

Documented uses

  • Save and retrieve semantic memories across agent sessions.
  • Represent entities and relationships, then search or traverse the knowledge graph.
  • Track changing facts, assess contradictions and allow unused memories to decay.
  • Expose memory functions to an MCP-based agent workflow.
Before you choose

Requirements & limits

The README describes self-hosted use and displays an Apache 2.0 license badge.

  • The documented agent setup uses an MCP server and exposes memory, graph, session and cognitive tools.
  • The project offers distinct integration paths: MCP, a Python REST client and an embedded Go engine.
  • The README says demo-mode data is temporary and discarded on exit; text search there needs an embedder, while vector search works out of the box.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up Kektordb as an MCP memory server for OpenCode

Configure the documented OpenCode MCP setup and start Kektordb with agent tools.

Before you start

  • Kektordb
  • An OpenCode setup

Steps

  1. Run kektordb setup opencode once to install and configure the OpenCode integration.

  2. Launch the memory server with kektordb --mcp --tools=agent.

Expected outcome

The README says the agent can call memory, retrieval and session tools, with memories persisting across sessions.

Guide limitations

  • The README documents this setup path but the supplied material does not provide further troubleshooting or framework compatibility details.
Documentation 1 ↗
COMPARE OPTIONS

Compare options for Vector search

These projects document a common task and share a project role. This is a source-based comparison, not an execution benchmark or a guarantee of interchangeability.

ProjectDocumented purposePlatformAccess and requirements
KektordbSave and retrieve semantic memories across agent sessions.Self-hosted, ServerThe README describes self-hosted use and displays an Apache 2.0 license badge.
LongbowStore and retrieve dense or sparse vectors with HNSW search.Server
Memento MCPAdd persistent knowledge-graph memory to an MCP-compatible client such as Claude Desktop, Cursor or GitHub Copilot.Cloud, Self-hosted, Server
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
17 October 2026
Original website title
sanonone/kektordb · Official homepage title · retained in its source language
How we review sourcesSuggest a correction