The DependencyUpgrade patchflow aims to automatically update vulnerable versions of dependencies in your repository to the fixed version.

​
How to run?

You can run it as follows:

patchwork DependencyUpgrade

by default you will need to provide the openai_api_key and the github_api_key you can pass them as arguments:

patchwork DependencyUpgrade openai_api_key=<Your_API_KEY> github_api_key=<Your_GH_Token>

​
What it does?

The DependencyUpgrade patchflow will first scan your repository using an open-source scanner dep-scan. It will then extract the vulnerable libraries information detected by the scan and use them to create a prompt to be sent to gpt-3.5-turbo to update your package manager file. You can check the default prompt template. The fixed package manager file is then committed to the repository under a new branch and finally a pull request is created for the user to review and merge the changes.

​
Configuration

The following are the default configurations that can be modified by the user to adapt the DependencyUpgrade patchflow to their needs. All the options can be set both via CLI arguments and and the yaml config file.

​
Model

You can choose any LLM API as long as it has an OpenAI API compatible chat completions endpoint. Just update the default values of the following options:

- model: gpt-3.5-turbo
- client_base_url: https://api.openai.com/v1

E.g. to use Meta’s CodeLlama model from HuggingFace you can set:

client_base_url: https://api-inference.huggingface.co/models/codellama/CodeLlama-70b-Instruct-hf/v1
model: codellama/CodeLlama-70b-Instruct-hf
model_temperature: 0.2
model_top_p: 0.95
model_max_tokens: 2000

and pass your HuggingFace token in the openai_api_key option.

You can also use llama.cpp to run inference on CPU locally. Just install the llama-cpp-python package and run their OpenAI compatible web server as described here with the command:

python3 -m llama_cpp.server --hf_model_repo_id TheBloke/deepseek-coder-6.7B-instruct-GGUF --model 'deepseek-coder-6.7b-instruct.Q4_0.gguf'

Once the local server is running you can set:

client_base_url: http://localhost:8000/v1
model: TheBloke/deepseek-coder-6.7B-instruct-GGUF
model_temperature: 0.2
model_top_p: 0.95
model_max_tokens: 1000

and use the local model for inference.

​
Upgrade Threshold

You can set the upgrade threshold you want to update the libraries to. The accepted values are ‘major’, ‘minor’ or ‘patch’. E.g.

upgrade_threshold: minor

would only upgrade across minor and patch versions of the library, i.e. an upgrade to another major wouldn’t be done.

​
Severity

You can also set the severity of the vulnerabilities that you want to upgrade. Severity is derived from the CVSS score and can take values of ‘none’, ‘low’, ‘medium’, ‘high’, and ‘critical’. E.g.

severity: medium

means that all libraries that have medium and above vulnerabilities will be upgraded.

​
Analyze Impact

By default we update the libraries to their fixed versions. If the upgrades are across major versions or a lot has changed between the versions of the library updating it may break your application. The analyze_impact option can be set to enable an impact analysis of the library upgrades as follows:

analyze_impact: true
libraries_api_key: <Your_API_KEY>

This will get the diff between the vulnerable and fixed versions of the library and determine the impacted methods. Then, we will analyze your repository to determine if you are using the impacted methods. And, finally we will use this information to prompt an LLM to migrate the code that uses these impacted methods in your repository. The code changes are then committed in a new branch in your repository to be included in the same PR that has the update t the package manager file. We use libraries.io as the source of information about libraries. You will need to get a free api key from them and set in the options.

​
Manage PRs

In addition, there are options to let you manage the PRs as you like, by setting a branch_prefix, or disabling the creation of new branches with disable_branch (commits will be made on the current branch). You can also disable PR creation with disable_pr or force push commits to existing PR with force_pr_creation.

branch_prefix: Auto-Generated-Dependency-Upgrades
disable_branch: false
disable_pr: false
force_pr_creation: false

​
Prompt template

You can update the default prompt template. The basic prompt that upgrades the dependencies is with "id": "depupgrade". Note the use of variables {{Updates}} and {{PackageManagerFile}}. They are generated by the steps within the DependencyUpgrade patchflow and replaced by the actual values during the execution. The expected output response is complete content of the package manager file with the libraries updated to their fixed versions.

​
Examples

Here are some example PRs generated with the DependencyUpgrade patchflow:

GenerateREADMEResolveIssue