OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Coding Tools

Agentless

Agentless is an open-source Python tool that uses an LLM-based process to localize software faults, generate candidate code patches and validate candidate fixes with regression and reproduction tests.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Agentless — OpenViST project profile in AI Coding Tools.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Agentless when you want a structured, test-informed workflow for investigating software issues and comparing candidate patches, including work on SWE-bench problems.

Who it is for

  • Developers investigating software issues and considering candidate repairs.
  • Researchers working with software issue-solving workflows such as SWE-bench.

Documented uses

  • Narrow a reported fault from relevant files to classes, functions and edit locations.
  • Generate multiple candidate patches for a software issue.
  • Use regression and reproduction test results to re-rank candidate fixes.
Before you choose

Requirements & limits

The project is publicly available as open-source software under the MIT license. Its setup instructions require an OpenAI API key.

  • The documented setup uses Conda, Python 3.11, the repository requirements file and an OpenAI API key.
  • Agentless describes a three-stage process: localization, repair and patch validation.
  • The README points to a separate guide for reproducing full SWE-bench Lite experiments.

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 Agentless for SWE-bench problems

Prepare the documented Python environment and API key so Agentless is ready to run on SWE-bench problems.

Before you start

  • Git
  • Conda
  • Python 3.11 environment
  • OpenAI API key

Steps

  1. Clone the Agentless repository and change into its directory.

  2. Create and activate a Conda environment using Python 3.11.

  3. Install dependencies with the repository’s requirements file and add the current directory to PYTHONPATH.

  4. Export the OpenAI API key as OPENAI_API_KEY.

Expected outcome

The README describes the environment as ready to run Agentless on SWE-bench problems.

Guide limitations

  • These steps cover environment setup; they do not specify a particular command for running a problem.
  • Reproducing the full SWE-bench Lite experiments requires following the separate README_swebench.md instructions.
Documentation 1 ↗
Evidence and transparency

Sources & review

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