---

Auto-Swe

An AI-powered automation system for processing Git repositories, generating PRs, and maintaining vectorized repo metadata for advanced analysis.

High-level architecture

Features

• Repo Indexing:

  • Automatically indexes repository metadata into PGVector.
  • Skips indexing if already present.

• Automated PR Generation:

  • Processes repositories using LangGraph nodes.
  • Creates Pull Requests based on AI-generated analysis or updates.

• Incremental Updates:

  • Watches push events to main or master branches.
  • Updates indexed data incrementally without reprocessing the whole repo.

• Queue & Worker System:

  • Uses BullMQ for background job management.
  • Supports scalable processing with multiple workers.

• Sandboxed Execution:

  • Runs code safely using E2B sandbox.

---

Tech Stack

• Language & Runtime: Bun
• Monorepo Management: Bun workspace
• Database: PostgreSQL + PGVector for embeddings
• Queue System: BullMQ
• Sandboxing: E2B
• AI Processing: LangGraph nodes for repo analysis

---

Getting Started

Prerequisites

• Bun installed (v1.x+)
• PostgreSQL with PGVector extension
• Redis (for BullMQ)
• Node-compatible environment for E2B sandbox

Installation

# Clone the monorepo
git clone https://github.com/hemanth-1321/auto-swe
cd auto-swe

# Install dependencies using Bun
bun install

Environment Variables

Create a .env file in the root with:

MY_GITHUB_APP_ID=
MY_GITHUB_CLIENT_ID=
MY_GITHUB_CLIENT_SECRET=
MY_GITHUB_PRIVATE_KEY_PATH="./private-key.pem"
MY_GITHUB_WEBHOOK_SECRET=
HF_TOKEN=
REDIS_URL=
GROQ_API_KEY=
E2B_API_KEY=
PG_URL=
OPENAI_API_KEY=
DATABASE_URL=
JWT_SECRET=

Folder Structure

/apps
  /worker         # BullMQ worker processing jobs
  /server            # API server for triggering repo jobs
  /web               #next js frontend
/packages
  /db             # PGVector integration
  /redis            #redis
  /shared        # github

---