OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Research Agents

Kestrelai

Kestrelai is a Python-based, local-first research assistant that schedules and executes evidence-gathering tasks over extended periods, using Ollama-backed language models and a dashboard for progress, provenance, checkpoints and reports.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Kestrelai — OpenViST project profile in Research Agents.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Kestrelai when a research task needs repeated evidence gathering and follow-up exploration over hours or days, rather than a single answer generated from one prompt.

Who it is for

  • Researchers handling long-running investigations
  • Developers exploring self-hosted research agents
  • Users who want small local language models in a research workflow
  • Teams building persistent literature-review or research-monitoring processes

Documented uses

  • Break a research question into concrete search, retrieval, filtering and extraction tasks
  • Continue research work in the background over a defined time-box
  • Track consulted searches and source evidence
  • Review checkpoints and redirect exploration
  • Assemble evidence for comparisons, reports or follow-up analysis
Before you choose

Requirements & limits

Kestrelai is presented as a public, MIT-licensed repository for local setup. The documented installation uses Docker Desktop and Ollama, with a browser interface launched locally.

  • The project is designed for small Ollama-served models of roughly 1–7B parameters.
  • Its documented components are FastAPI, React, Redis and Ollama.
  • The workflow exposes searches, sources, checkpoints and reports for review.
  • Docker Desktop is part of the documented setup, while Ollama may be supplied through the Dockerized option.
  • The repository describes the project as under construction.

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 Kestrelai locally with Docker and Ollama

Build and launch the documented local Kestrelai application.

Before you start

  • Docker Desktop
  • Ollama, unless using the Dockerized version
  • A cloned copy of the public Kestrelai repository

Steps

  1. Clone the Kestrelai repository from GitHub.

  2. Ensure Docker Desktop is running.

  3. From the repository root, run the documented Docker Compose build and launch command.

  4. Open the local web interface at the documented default address.

  5. Create a research task with a description and time-box, then use the dashboard to monitor searches, sources, checkpoints, reports and exports.

Expected outcome

A locally launched Kestrelai interface for creating and monitoring research tasks.

Guide limitations

  • The repository describes the project as under construction.
  • The setup documentation describes local launch and dashboard use; it does not establish compatibility beyond the listed environments and prerequisites.
Documentation 1 ↗
Evidence and transparency

Sources & review

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