https://docs.sweep.dev/assistant

[banner]HomeHomeEmailEmail (opens in a new tab)Sweep ProSweep Pro
(opens in a new tab)
[                    ]
 GitHub (opens in a new tab) Discord (opens in a new tab)
[                    ]

  * What's Sweep?
  *  Getting Started
  *  Sweep Assistant
  *  Self-Host
  *  Usage
      + Tutorial: Solving an Example Ticket with Sweep
      + Advanced Sweep Features: becoming a Power User
      + Config: Customize Sweep using sweep.yaml
      + Extra Self Hosting Configurations
      + Sandbox: Giving Sweep Dev Tools
  *  About
      + Examples
      + Limitations
      + Roadmap
  *  FAQ
  *  Blogs
      +  Releasing our Open Source LLM Framework + Playground!
      +  Large Language Models are Bad at Refactoring Code
      + [?] A Simple Proxy for Azure and OpenAI raised our GPT4 TPM
        limit by 24x
      +  Having GPT-4 Iterate on Unit Tests like a Human
      +  Zero Downtime Deployment with < 50 lines of bash
      + [?] How an AI plans code changes across an entire GitHub
        Repository
      + [?] AI Junior Dev meets Super-Linter
      +  Why getting GPT-4 to modify files is hard
      +  When you should handle tech debt
      +  Self Hosting an AI Junior Developer
      +  Improving LlamaIndex's Code Chunker by Cleaning Tree-Sitter
        CSTs
      +  How our AI junior dev reads all of your documentation
      +  Sweep's Core Algorithm
      + [?][?] Generating 50k+ embeddings in 25 seconds using GTE  and
        MapReduce 
      +  Chunking 2M+ files a day for Code Search using Syntax Trees
      +  Letting an AI Junior Dev run GitHub Actions
      +  Code Search Infra for an AI junior developer - that doesn't
        store code
      +  Understanding your codebase with ctags
      + [?] Building a code search engine in one day
      +  GPT 32k,  Open-source, DeepLake [?], GPT Functions Search 
        and more!
      + Index
  *  Videos
      +  Introducing Sweep
      + [?] Adding a Banner to our Landing Page using Sweep
      +  AI Junior Dev Refactors Itself
      +  Adding Walrus Operator to our Code using Sweep
      +  AI junior developer adds an easter egg to our logo
  *  Privacy Policy

On This Page

  * Installation (2 min)
  * Usage (After Installation)
  * Set Repository Path
  * Select a files for Sweep to edit
  * Add Instructions
  * Generate Code
  * Checking your Code

Question? Give us feedback - (opens in a new tab)Edit this page
 Sweep Assistant

Sweep Assistant

Sweep is a locally running dev tool that codes alongside you. To
preserve your security, Sweep will only send code to your OpenAI
endpoint. If you use Azure OpenAI, all of your code will stay within
your network.

Sweep Assistant

Installation (2 min) 

Prerequisites

     1. OpenAI API key(https://platform.openai.com (opens in a new
        tab))
     2. Node.js v18+(https://nodejs.org/en/download/ (opens in a new
        tab)). Use npm install -g n && n 18 to use Node.js v18.

Install Sweep by running the following command in your terminal:

npm (recommended)curl
terminal

npm i -g sweepai && npx sweepai build
npx sweepai

Having Trouble

 1. If you see a build error, you may be on version 16 of Node.js.
    You can upgrade to version 18 by running the following command:

terminal

npm install -g n && n 18

 2. If you see a permissions issue, you can fix it by running the
    following command:

terminal

sudo (npm i -g sweepai && npx sweepai build)

You can run npx sweepai to start Sweep in the future. To update the
assistant, run

terminal

npm update -g sweepai && npx sweepai build

Disabling Telemetry

We collect telemetry (not code) to improve the assistant, such as the
types of tasks you're asking Sweep to handle.

To disable telemetry, add NEXT_PUBLIC_NO_TELEMETRY=true to your .env
in your Sweep Assistant directory.

You can find where the Sweep directory is cloned by running which
sweep. For example, if which sweep yields sweep: aliased to npm start
--prefix /home/bob/sweep/platform, then you would run:

terminal

cd /home/bob/sweep
echo "NEXT_PUBLIC_NO_TELEMETRY=true" >> platform/.env

Usage (After Installation) 

Set Repository Path 

You can run pwd to use your current working directory.

Select a files for Sweep to edit 

Click "Modify file" or "Create file".

Add Instructions 

Add meticulous instructions for the code to edit. You can type @ to
mention a file.

Generate Code 

Click "Generate Code" to run the assistant. Then click the check mark
to approve the change or the reload to reject the change.

Checking your Code 

We highly recommended that you add a checks to your assistant. There
are two types of checks:

 1. Validation Script: runs after every edit. We recommend using a
    formatter and linter to catch syntax errors and undefined
    variables.
 2. Test Suite: runs after all edits are complete.

Why do we need a validation script?

Human developers identify syntax errors using visual cues like syntax
highlighting and red squiggles, which is unavailable to LLMs. The
best way for LLMs to detect such errors is to use a formatter to
check the syntax and a linter for catching undefined variables.

Here are some example scripts we recommend:

PythonJavascript / Typescript
Validation Script

python3 -m py_compile < $FILE_PATH
pylint $FILE_PATH --errors-only

Test Suite

pytest $FILE_PATH

Make sure to click the "Run Tests" button to test the script.

You may want to provide the full path to the CLI tools (e.g. /home/
kevin/.cache/pypoetry/virtualenvs/sweepai-u_abcdefg-py3.10/bin/black)
if you are using a custom virtual environment. You may also want to
run python -m pylint or python3 -m pylint instead of pylint.

 Getting Started Self-Host
---------------------------------------------------------------------
Sweep AI (c) 2024