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

Ataru

Ataru is a local conversation indexing and retrieval application for AI-agent histories. It supports Claude and Codex sources and provides a Rust search core through an SDK, API and JSON CLI, alongside a desktop GUI. Keyword search uses Tantivy and Jieba; semantic recall is optional. Results can be organized by turn, run, session or project and linked back to original conversation context.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Ataru — 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 Ataru when relevant work is already recorded in supported local agent histories and you need to recover it as reusable context.

Who it is for

  • Developers who need to find earlier answers, commands, troubleshooting steps or decisions in local AI-agent histories.
  • Teams or tool builders seeking a shared search interface for indexed agent conversations through a GUI, CLI, API or SDK.

Documented uses

  • Search local Claude and Codex conversation histories for a remembered detail.
  • Retrieve context by turn, run, session or project rather than browsing files manually.
  • Return from a search match to its original message and surrounding context.
  • Use keyword retrieval or optional semantic recall for less exact queries.
Before you choose

Requirements & limits

The project lists desktop builds for macOS, Windows and Linux and publishes its code under the Apache-2.0 license.

  • The documented session sources include Claude and Codex; support for another source is an experimental adaptation, not a documented connector.
  • Keyword search is available through the JSON CLI Skills; semantic or hybrid retrieval is described for the desktop app.
  • The first desktop search may require building or updating the local index before queries can run.
  • The search core and desktop interface are separate access surfaces, with CLI and API/SDK options for integrations.

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 Ataru’s conversation-search Agent Skill

Install the published indexing and search Skills for retrieving prior session context through Ataru’s JSON CLI.

Before you start

  • An environment that can run the documented npx command.
  • Ataru JSON CLI version 0.41.3 or later; the Skills perform a version check.

Steps

  1. Install the indexing Skill with the command below.

    Documented command
    npx lovstudio skills add ataru-indexing

  2. Install the search Skill with the command below.

    Documented command
    npx lovstudio skills add ataru-search

  3. Ensure the index is ready before searching; the search Skill refuses to run when the index is not ready.

  4. Use the Skill to search and inspect returned context using stable IDs.

Expected outcome

The indexing and search Skills are installed; search results can include locatable prior conversation context when the index is ready.

Guide limitations

  • The JSON CLI supports keyword mode only; semantic or hybrid retrieval requires the desktop app.
  • The Skills do not maintain a separate index, and the search Skill reports an error rather than treating an unavailable index as zero matches.
  • The documented large-corpus turn-level query took about 48 seconds; the Skills use a 180-second default timeout and 300 seconds for reads.
Documentation 1 ↗
Evidence and transparency

Sources & review

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