huggingface/serge
Reviews pull requests with any OpenAI-compatible LLM
https://skillcdn.ai/gh/huggingface/sergeConnect this address to your AI to use these skills. How to connect
- Unverified
- Default branch
main - Commit
02333f4 - License
Apache-2.0
Explore this repository
Searches the original text of skills and documents; display translations are not searched.
Folders
Documents
- _config.ymldocs/_config.yml
- ArchitectureThis page explains how serge is put together — its components, the execution backends that run the LLM work, and the two deployment flavors (Docker and Kubernetes). For the step-by-step review lifecy…docs/architecture.md
- HF inference costsThe task and review pages display the latest HF-reported inference cost when the configured HF key can read billing for the account being charged. This includes web reviews, webhook reviews, and inte…docs/billing.md
- ConfigurationConfiguration is passed as Action inputs in GitHub Action mode and as environment variables in server modes.docs/configuration.md
- DevelopmentThe tests cover trigger gating, config parsing, LLM response handling, review publishing, context scripts, helper tools, clone cache behavior, web app webhooks, provider configs, and persistence.docs/development.md
- Getting StartedThe quickest way to run serge is the GitHub Action. It needs an LLM API key, a workflow file, and a trigger comment on an open pull request.docs/getting-started.md
- GitHub ActionThe GitHub Action runs inside the target repository's workflow. It is the lowest-friction way to try serge because it does not require a server or GitHub App.docs/github-action.md
- GitHub App WebhookGitHub App mode runs a webhook server that listens for trigger comments, calls the LLM, and posts reviews with a GitHub App installation token. It is useful when you want one hosted reviewer to serve…docs/github-app.md
- How It WorksA review starts from an ordinary PR comment that begins with the trigger (@askserge must be the first word):docs/how-it-works.md
- sergeserge reviews pull requests with an OpenAI-compatible LLM and posts inline comments through GitHub's Pull Request Reviews API. The default persona is Serge, and the default trigger is @askserge.docs/index.md
- LLM Providersserge talks to OpenAI-compatible chat completion endpoints. The endpoint should support:docs/llm-providers.md
- Repository CustomizationRepositories can shape reviews with files on their default branch. Reading customization from the default branch prevents PR authors from changing review policy in the same PR being reviewed.docs/repository-customization.md
- Security architecture: isolating PR/fork code from the review hostai-reviewer reviews pull requests, including PRs from forks opened by untrusted authors. The persistent review host (an AWS EC2 box running Amazon Linux 2023 as ec2-user under systemd) holds high-val…docs/security-architecture.md
- Securityserge treats PR content as untrusted input.docs/security.md
- Tasks flow (write-capable)The tasks flow lets a GitHub Actions job ask serge to produce a change on the repository — open a pull request with a fix, or push a follow-up commit onto an existing serge-authored fix branch. It is…docs/tasks-flow.md
- Normalize validation (in-loop)When serge produces a fix through the tasks flow, the LLM proposes a logical patch — it does not run the target repo's build commands. For repos with a consistency gate (regenerated files, formatting…docs/tasks-normalize.md
- TroubleshootingCheck that:docs/troubleshooting.md
- Web app tourA screen-by-screen tour of the serge web app (reviewbot-web). For setup and the trust model behind staged reviews, see Staged web app; for how the whole thing fits together, see Architecture.docs/web-app-tour.md
- Staged Web AppThe web app lets a signed-in reviewer start a review, watch the LLM stream its draft, edit the summary and inline comments, discard noisy comments, and only then publish the review to GitHub.docs/web-app.md