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

KITE

KITE is an open-source Python library that gives AI agents long-term memory through structured, topic-indexed conversational facts and query-based retrieval. It avoids embeddings and vector databases, compiles questions into inspectable plans, and can return facts with links to their source statements.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

KITE — 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 KITE when an agent needs conversational recall that can account for dates and updates, expose how it selected facts, and retain the original statements behind an answer.

Who it is for

  • Developers building AI agents that need to retain and retrieve conversation details.
  • Teams evaluating vector-free memory on long-conversation benchmarks.

Documented uses

  • Convert conversation messages into structured, source-backed facts.
  • Retrieve relevant facts for a question using an inspectable query plan.
  • Produce answers with evidence pointing back to the conversation.
  • Evaluate memory approaches against LoCoMo and LongMemEval-S.
Before you choose

Requirements & limits

The project is published as open source under the Apache-2.0 license. Its README quick start requires an OpenAI API key; it also documents an OPENAI_BASE_URL setting for another OpenAI-compatible provider.

  • KITE is a Python library, with a CLI platform classification; it is a component for agent builders rather than a complete hosted agent application.
  • The documented retrieval approach uses topic-indexed facts and symbolic plans instead of vector search.
  • The README’s quick start depends on an OpenAI API key, or a configured OpenAI-compatible provider.
  • The repository reports benchmark scores of 93.51% on LoCoMo and 85.60% on LongMemEval-S, using gpt-4.1-mini across benchmark stages; results are project-reported.

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

Install KITE and configure its documented quick-start inputs

Set up the Python package, provider credential, and example XML file used in the README quick start.

Before you start

  • Python and pip
  • An OpenAI API key, set as OPENAI_API_KEY
  • For another OpenAI-compatible provider, an OPENAI_BASE_URL value

Steps

  1. Install the package with the command below.

    Documented command
    python -m pip install memoket-kite

  2. Set OPENAI_API_KEY in the environment.

  3. Create an artifacts directory and download the README’s demo_codebook.xml to artifacts/quickstart.xml.

Expected outcome

The documented package, provider credential, and example codebook are in place for the README’s subsequent remember and recall examples.

Guide limitations

  • This guide covers setup only; it does not claim that the commands were executed or verify provider access.
  • The README’s remember, recall, and answer examples are separate usage stages.
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
KITEConvert conversation messages into structured, source-backed facts.CLIThe project is published as open source under the Apache-2.0 license. Its README quick start requires an OpenAI API key; it also documents an OPENAI_BASE_URL setting for another OpenAI-compatible provider.
Engram MemoryPersist agent memories across sessions.Linux, macOS, Self-hosted, WindowsThe repository README identifies a BSL 1.1 license. It describes the Community edition as free for internal use, research and self-hosted deployments; commercial or SaaS use requires a license, and each version converts to Apache 2.0 after four years. The self-hosted setup requires Docker, 4 GB or more of RAM and 10 GB or more of storage. Python 3.10 or later is needed only to run the stdio MCP server or CLI tools di
YourmemoryRetain durable facts and preferences across agent sessions.CLI, Self-hosted, ServerThe Python package setup requires obtaining a token through the project website by entering an email and verifying a six-digit code. The README also describes standalone binaries for macOS, Linux and Windows.
Evidence and transparency

Sources & review

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