CodiumAI PR-Agent aims to help efficiently review and handle pull requests, by providing AI feedbacks and suggestions
-
See the Installation Guide for instructions on installing PR-Agent on different platforms.
-
See the Usage Guide for instructions on running PR-Agent tools via different interfaces, such as CLI, PR Comments, or by automatically triggering them when a new PR is opened.
-
See the Tools Guide for a detailed description of the different tools, and the available configurations for each tool.
- News and Updates
- Overview
- Example results
- Try it now
- PR-Agent Pro ๐
- How it works
- Why use PR-Agent?
We released a new version of PR-Agent Chrome extension, with two new features:
- PR-Agent filters ๐จ
- Code suggestions interactions ๐
See more here
Check out CodiumAI new project, Cover-Agent, that can automatically generate qualified tests to enhance existing test suites, aiming to increase code and behavior coverage efficiently.
GPT-4o is now the default fast model ("Turbo"). This model will be used for all commands except review
and improve
, which will still use "GPT-4-2024-04-09", since they are harder and would still benefit from the larger model.
Inspired by AlphaCodium flow engineering scheme, PR-Agent now performs self-reflection on the code suggestions it provides, enabling to remove invalid suggestions, and score the valid ones. The suggestions will be presented sorted by their score, enabling to focus on the most important ones first.
You can also choose to automatically remove suggestions below a certain importance score threshold, by setting the pr_code_suggestions.suggestions_score_threshold
configuration.
Supported commands per platform:
GitHub | Gitlab | Bitbucket | Azure DevOps | ||
---|---|---|---|---|---|
TOOLS | Review | โ | โ | โ | โ |
โฎ Incremental | โ | ||||
โฎ SOC2 Compliance ๐ | โ | โ | โ | โ | |
Describe | โ | โ | โ | โ | |
โฎ Inline File Summary ๐ | โ | ||||
Improve | โ | โ | โ | โ | |
โฎ Extended | โ | โ | โ | โ | |
Ask | โ | โ | โ | โ | |
โฎ Ask on code lines | โ | โ | |||
Custom Prompt ๐ | โ | โ | โ | โ | |
Test ๐ | โ | โ | โ | ||
Reflect and Review | โ | โ | โ | โ | |
Update CHANGELOG.md | โ | โ | โ | โ | |
Find Similar Issue | โ | ||||
Add PR Documentation ๐ | โ | โ | โ | ||
Custom Labels ๐ | โ | โ | โ | ||
Analyze ๐ | โ | โ | โ | ||
CI Feedback ๐ | โ | ||||
Similar Code ๐ | โ | ||||
USAGE | CLI | โ | โ | โ | โ |
App / webhook | โ | โ | โ | โ | |
Tagging bot | โ | ||||
Actions | โ | โ | |||
CORE | PR compression | โ | โ | โ | โ |
Repo language prioritization | โ | โ | โ | โ | |
Adaptive and token-aware file patch fitting | โ | โ | โ | โ | |
Multiple models support | โ | โ | โ | โ | |
Static code analysis ๐ | โ | โ | โ | โ | |
Global and wiki configurations ๐ | โ | โ | โ | โ | |
PR interactive actions ๐ | โ |
- ๐ means this feature is available only in PR-Agent Pro
โฃ Auto Description (/describe
): Automatically generating PR description - title, type, summary, code walkthrough and labels.
โฃ Auto Review (/review
): Adjustable feedback about the PR, possible issues, security concerns, review effort and more.
โฃ Code Suggestions (/improve
): Code suggestions for improving the PR.
โฃ Question Answering (/ask ...
): Answering free-text questions about the PR.
โฃ Update Changelog (/update_changelog
): Automatically updating the CHANGELOG.md file with the PR changes.
โฃ Find Similar Issue (/similar_issue
): Automatically retrieves and presents similar issues.
โฃ Add Documentation ๐ (/add_docs
): Generates documentation to methods/functions/classes that changed in the PR.
โฃ Generate Custom Labels ๐ (/generate_labels
): Generates custom labels for the PR, based on specific guidelines defined by the user.
โฃ Analyze ๐ (/analyze
): Identify code components that changed in the PR, and enables to interactively generate tests, docs, and code suggestions for each component.
โฃ Custom Prompt ๐ (/custom_prompt
): Automatically generates custom suggestions for improving the PR code, based on specific guidelines defined by the user.
โฃ Generate Tests ๐ (/test component_name
): Generates unit tests for a selected component, based on the PR code changes.
โฃ CI Feedback ๐ (/checks ci_job
): Automatically generates feedback and analysis for a failed CI job.
โฃ Similar Code ๐ (/find_similar_component
): Retrieves the most similar code components from inside the organization's codebase, or from open-source code.
Try the GPT-4 powered PR-Agent instantly on your public GitHub repository. Just mention @CodiumAI-Agent
and add the desired command in any PR comment. The agent will generate a response based on your command.
For example, add a comment to any pull request with the following text:
@CodiumAI-Agent /review
and the agent will respond with a review of your PR
To set up your own PR-Agent, see the Installation section below.
Note that when you set your own PR-Agent or use CodiumAI hosted PR-Agent, there is no need to mention @CodiumAI-Agent ...
. Instead, directly start with the command, e.g., /ask ...
.
PR-Agent Pro is a hosted version of PR-Agent, provided by CodiumAI. It is available for a monthly fee, and provides the following benefits:
- Fully managed - We take care of everything for you - hosting, models, regular updates, and more. Installation is as simple as signing up and adding the PR-Agent app to your GitHub\GitLab\BitBucket repo.
- Improved privacy - No data will be stored or used to train models. PR-Agent Pro will employ zero data retention, and will use an OpenAI account with zero data retention.
- Improved support - PR-Agent Pro users will receive priority support, and will be able to request new features and capabilities.
- Extra features -In addition to the benefits listed above, PR-Agent Pro will emphasize more customization, and the usage of static code analysis, in addition to LLM logic, to improve results. See here for a list of features available in PR-Agent Pro.
The following diagram illustrates PR-Agent tools and their flow:
Check out the PR Compression strategy page for more details on how we convert a code diff to a manageable LLM prompt
A reasonable question that can be asked is: "Why use PR-Agent? What makes it stand out from existing tools?"
Here are some advantages of PR-Agent:
- We emphasize real-life practical usage. Each tool (review, improve, ask, ...) has a single GPT-4 call, no more. We feel that this is critical for realistic team usage - obtaining an answer quickly (~30 seconds) and affordably.
- Our PR Compression strategy is a core ability that enables to effectively tackle both short and long PRs.
- Our JSON prompting strategy enables to have modular, customizable tools. For example, the '/review' tool categories can be controlled via the configuration file. Adding additional categories is easy and accessible.
- We support multiple git providers (GitHub, Gitlab, Bitbucket), multiple ways to use the tool (CLI, GitHub Action, GitHub App, Docker, ...), and multiple models (GPT-4, GPT-3.5, Anthropic, Cohere, Llama2).
- If you host PR-Agent with your OpenAI API key, it is between you and OpenAI. You can read their API data privacy policy here: https://openai.com/enterprise-privacy
-
When using PR-Agent Pro ๐, hosted by CodiumAI, we will not store any of your data, nor will we use it for training. You will also benefit from an OpenAI account with zero data retention.
-
For certain clients, CodiumAI-hosted PR-Agent Pro will use CodiumAIโs proprietary models โ if this is the case, you will be notified.
-
No passive collection of Code and Pull Requestsโ data โ PR-Agent will be active only when you invoke it, and it will then extract and analyze only data relevant to the executed command and queried pull request.
- The PR-Agent Chrome extension serves solely to modify the visual appearance of a GitHub PR screen. It does not transmit any user's repo or pull request code. Code is only sent for processing when a user submits a GitHub comment that activates a PR-Agent tool, in accordance with the standard privacy policy of PR-Agent.
- Discord community: https://discord.gg/kG35uSHDBc
- CodiumAI site: https://codium.ai
- Blog: https://www.codium.ai/blog/
- Troubleshooting: https://www.codium.ai/blog/technical-faq-and-troubleshooting/
- Support: support@codium.ai