Skip to content

README.md

Latest commit

 

History

History
110 lines (83 loc) · 6.39 KB

README.md

File metadata and controls

110 lines (83 loc) · 6.39 KB
layout default
title Sweep Tutorial
nav_order 182
has_children true
format_version v2

Sweep Tutorial: Issue-to-PR AI Coding Workflows on GitHub

Learn how to use sweepai/sweep to turn GitHub issues into pull requests, operate feedback loops, and run self-hosted or CLI workflows with clear guardrails.

GitHub Repo License Docs Website

Product Status Notice: The original Sweep GitHub issue-to-PR product has been abandoned. The sweepai/sweep README now redirects users to a new JetBrains AI coding plugin. The hosted Sweep GitHub App and the issue-to-PR workflow documented in this tutorial are no longer actively supported. This tutorial is preserved as a reference for the architecture patterns and operational workflows Sweep introduced. Teams building new automation should evaluate current alternatives such as SWE-agent, Aider, or OpenHands.

Why This Track Matters

Sweep popularized an issue-to-PR coding-agent workflow on GitHub. The repository and docs still provide strong reference patterns for asynchronous AI delivery loops, feedback handling, and operational controls even though the original hosted product is no longer active.

This track focuses on:

  • understanding the original Sweep issue-to-PR architecture and config model
  • configuring repository-level behavior through sweep.yaml
  • operating review, CI, and retry loops for higher output quality
  • understanding self-hosted and local CLI deployment paths as legacy reference

Current Snapshot (auto-updated)

Mental Model

flowchart LR
    A[GitHub issue or comment] --> B[Sweep task intake]
    B --> C[Codebase search and planning]
    C --> D[PR creation and code edits]
    D --> E[CI feedback and retries]
    E --> F[Human review and merge]
Loading

Chapter Guide

Chapter Key Question Outcome
01 - Getting Started and Current Product Posture Which Sweep path should teams use today? Clear starting point
02 - Issue to PR Workflow Architecture How does Sweep move from issue text to code changes? Better task decomposition
03 - Repository Configuration and Governance How do sweep.yaml controls shape behavior? Stronger repo policy
04 - Feedback Loops, Review Comments, and CI Repair How do you iteratively improve generated PRs? Higher merge quality
05 - CLI and Self-Hosted Deployment When should teams run local CLI or self-hosted app paths? Deployment strategy
06 - Search, Planning, and Execution Patterns What internal workflow patterns matter most? Better reliability expectations
07 - Limitations, Risk Controls, and Safe Scope What constraints must teams respect? Fewer failed runs
08 - Migration Strategy and Long-Term Operations How do teams evolve workflow investments as products shift? Durable adoption plan

What You Will Learn

  • how to run Sweep with realistic task sizing and expectations
  • how to tune config and review loops for better issue-to-PR outcomes
  • how to choose between hosted, CLI, and self-hosted deployment modes
  • how to manage lifecycle and migration risk as agent tooling evolves

Source References

Related Tutorials

Start with Chapter 1: Getting Started and Current Product Posture.

Navigation & Backlinks

Full Chapter Map

  1. Chapter 1: Getting Started and Current Product Posture
  2. Chapter 2: Issue to PR Workflow Architecture
  3. Chapter 3: Repository Configuration and Governance
  4. Chapter 4: Feedback Loops, Review Comments, and CI Repair
  5. Chapter 5: CLI and Self-Hosted Deployment
  6. Chapter 6: Search, Planning, and Execution Patterns
  7. Chapter 7: Limitations, Risk Controls, and Safe Scope
  8. Chapter 8: Migration Strategy and Long-Term Operations

Generated by AI Codebase Knowledge Builder