Open-source workflow automation for dev teams

Customizable AI workflows to automate code reviews, docs and patches
Backed by
Combinator

Run and build intelligent patchflows

Use ready-to-go patchflows or create your own to accelerate mundane development tasks.

GenerateREADME patchflow aims to automatically generate documentation for a folder in your repository in form of a README.md file.

Command

patchwork GenerateREADME

Configuration

disable_branch
disable_pr
force_pr_creation

PRReview aims to automatically summarize and comment on PRs based on the changes made to the code repository.

Command

patchwork PRReview

Configuration

diff_summary
diff_suggestion
model
client_base_url

DependencyUpgrade patchflow aims to automatically update vulnerable versions of dependencies in your repository to the fixed version while mitigating breaking changes.

Command

patchwork DependencyUpgrade

Configuration

model
client_base_url
analyze_impact
disable_branch
disable_pr
force_pr_creation

Create your own patchflows using available steps and prompts - or write one from scratch if you need. You can also use the patchflow generator to create one from the patched app without writing a single line of code.

Command

patchwork YourOwnPatchflow

Integrate LLMs seamlessly into your SDLC

21+
Patchwork Integrations

Patchwork readily integrates with your existing development and security solutions allowing you to create complex, multi-step flows with ease.

Build the AI assistance you need.
After all you're a dev, not a pilot.

Fully Customizable

Customize our patchflows or create new ones in minutes via code or our no-code builder, it's your call.

Privacy First

Deploy patchwork within your own infrastructure with your own LLM API keys for complete privacy.

No code, Low code and All Code

Generate patchflows using text prompts, visual builders or real python code - your choice!

Free and Open Source

Patchwork is an open-source project and can be used for free, forever!

Blogs

Learn what's new and get a behind the scenes look at Patched

Static Analysis Evaluation Benchmark with OpenAI's GPT-4o Fine-Tuning
August 24, 2024
OpenAI invited Patched to fine-tune gpt-4o to build the SOTA model for this benchmark. We worked alongside their team to understand how performance can be improved by fine-tuning frontier models.
Comparative study of fine-tuning GPT-4o-mini, Gemini Flash 1.5 and Llama-3.1-8B
August 13, 2024
We compare fine-tuning GPT-4o-mini, Gemini Flash 1.5, and Llama-3.1-8B models using a custom vulnerability fixes dataset, with GPT-4o-mini showing the most significant improvement and setting a new benchmark.
MOA: Optimizing inference for diverse software development tasks
August 2, 2024
We discuss Mixture of Agents, an inference optimization technique that enhances gpt-4o-mini to outperform gpt-4-turbo at ~1/50th the cost for a variety of software development tasks.

Frequently Asked Questions

Know more about the open-source Patchwork framework, and our managed service Patched.

What exactly is Patchwork?
How is it different from Copilot?
How do I try it out?
Is Patchwork free?
What is Patched?
I have more questions. Who do I talk to?