# Greptile

> Greptile is an AI code review agent for GitHub and GitLab. It reviews pull requests with whole-repo context (not just diffs), produces actionable feedback, and adapts to team preferences through configurable standards and feedback-driven learning.

If evaluating: Key features → Benchmarks → Security → Pricing.  
If implementing: Quickstart → GitHub/GitLab integration → Customization overview → Troubleshooting.

## Start here

- [What is Greptile?](https://www.greptile.com/docs/introduction): Concepts + mental model: repository graph indexing, PR review flow, and feedback-driven learning.
- [5-minute quickstart](https://www.greptile.com/docs/quickstart): Fast setup (GitHub App or GitLab token+webhook) and run your first automated PR review.
- [Key features](https://www.greptile.com/docs/code-review/key-features): Differentiators overview: repository graph context, high-signal findings (not nitpicks), learning via feedback, MCP IDE workflows, and enterprise/self-host deployment.
- [Anatomy of a review](https://www.greptile.com/docs/code-review/first-pr-review): What Greptile posts on PRs and how to interpret each component (summary, confidence, diagrams, inline comments).

## Evaluate

- [Benchmarks](https://www.greptile.com/benchmarks): Methodology + results comparing 5 tools on 50 real bug PRs with default settings; links to reproducible sources and protocol.
- [Examples](https://www.greptile.com/examples): Public PR links showing real Greptile summaries and links to PRs with Greptile comments on popular repos.
- [Customers](https://www.greptile.com/customers): Proof and outcomes: quotes + case studies across company sizes and workflows.
- [Pricing](https://www.greptile.com/pricing): Plans, billing mechanics, 14 day free trial, and FAQs (review bot seats, contracts, chat/API pricing).
- [Brex case study](https://www.greptile.com/customers/brex): Brex Q&A on rollout, adoption, and how Greptile fits a large production workflow.

## How Greptile works

- [Graph-based codebase context](https://www.greptile.com/docs/how-greptile-works/graph-based-codebase-context): How Greptile builds and uses a repository graph to reason beyond the diff.
- [Memory and learning](https://www.greptile.com/docs/how-greptile-works/memory-and-learning): How PR comments/replies and thumbs up/down reactions tune future reviews and reduce noise over time.
- [Reducing nitpicks](https://www.greptile.com/docs/how-greptile-works/nitpicks): How Greptile suppresses low-impact feedback and focuses on what teams actually act on.
- [Custom rules](https://www.greptile.com/docs/how-greptile-works/custom-rules): What you can enforce (architecture/security/performance/standards) and how rules show up in reviews.

## Implement

- [GitHub/GitLab integration](https://www.greptile.com/docs/integrations/github-gitlab-integration): Connect providers and ensure reviews trigger reliably (GitHub App; GitLab token+webhooks; repo enablement).
- [Developer essentials](https://www.greptile.com/docs/code-review/developer-essentials): Day-to-day workflow: trigger with `@greptileai`, ask targeted checks, review drafts, train via thumbs up/down and replies.
- [Tips & recipes](https://www.greptile.com/docs/code-review/tips-recipes): Playbooks: draft PR “local review”, RFC/doc review, partial reviews, and high-signal prompt patterns.
- [Troubleshooting](https://www.greptile.com/docs/troubleshooting/common-issues): Fix common failures: config not applying, reviews not running, webhook/filter issues, slow/stuck reviews.
- [Deployment options](https://www.greptile.com/docs/deployment-options): Decision guide for cloud vs self-host, plus Docker Compose vs Kubernetes and the supported setup paths.

## Configure reviews

- [Customization overview](https://www.greptile.com/docs/code-review/customization-overview): Where configuration lives (Dashboard vs repo `greptile.json`) and how precedence/scope works.
- [greptile.json reference](https://www.greptile.com/docs/code-review/greptile-json-reference): Full config schema; must be in repo root and read from PR source branch; overrides dashboard settings (strictness, comment types, filters, triggers, model).
- [Controlling nitpickiness](https://www.greptile.com/docs/code-review/controlling-nitpickiness): Tune strictness, comment types, ignore patterns, and trigger behavior to control signal vs noise.
- [Custom standards & rules](https://www.greptile.com/docs/code-review/custom-standards): Add enforceable standards (rules + style-guide files), scope them, verify they apply, debug misses.
- [Training the learning system](https://www.greptile.com/docs/code-review/training-the-learning-system): How to train Greptile with thumbs up/down reactions and short explanations; what each signal teaches.
- [Pattern repositories](https://www.greptile.com/docs/code-review/pattern-repositories): Reference related repos (shared libs/docs/services) during review via `patternRepositories` for cross-repo context.

## Advanced use cases

- [MCP overview](https://www.greptile.com/docs/mcp-v2/overview): Use Greptile from IDE/agents: fetch PR feedback, apply fixes, manage standards, check status.
- [IDE setup](https://www.greptile.com/docs/mcp-v2/setup): Configure Greptile MCP in Cursor/Claude/VS Code and verify tools are enabled.
- [MCP custom context](https://www.greptile.com/docs/mcp-v2/custom-context): Manage coding standards from your IDE: list, search, and create custom-context patterns via MCP.
- [MCP tools reference](https://www.greptile.com/docs/mcp-v2/tools): Complete API reference for MCP tools, including repository-parameter requirements and request/response shapes.
- [API introduction](https://www.greptile.com/docs/api-reference/introduction): API overview: index repos, query/search in natural language, and integrate with Slack/Sentry/GitHub workflows.
- [Webhooks](https://www.greptile.com/docs/api-reference/webhooks): Verify webhooks and parse `sessionId`, `sources`, `statusEndpoint`; troubleshoot common review failures.

## Security and legal

- [Security](https://www.greptile.com/security): Security controls + hosting model, inference/data handling, logging/storage options, and links to related legal documents.
- [Privacy policy](https://www.greptile.com/security/privacy): Privacy notice for services + site: what Personal Information is processed, how it’s shared, and user rights.
- [Subprocessors](https://www.greptile.com/security/subprocessors): Subprocessor list (purpose + location) and how changes are communicated.
- [Terms of service](https://www.greptile.com/terms-of-service): Terms governing use of Greptile services, including data safeguards and processing provisions.

## Content (selected)

- [Best AI code review tools](https://www.greptile.com/content-library/best-ai-code-review-tools): Explains how AI code review works, tradeoffs, and a shortlist of top tools plus setup best practices.
- [Software Needs An Independent Auditor](https://www.greptile.com/blog/auditor): Why AI-generated code needs an independent PR “auditor” and why Greptile is review-only (it doesn’t generate code).
- [Greptile v3, an agentic approach to code review](https://www.greptile.com/blog/greptile-v3-agentic-code-review): Deep dive on the v3 agentic workflow and why it drives higher action rate and higher signal-to-noise than v2.
- [AI Code Review: Should the Author Be The Reviewer?](https://www.greptile.com/blog/ai-code-reviews-conflict): When your author is an AI, your reviewer should be independent; explains the incentives and why AI code needs closer review.

## Optional

- [Greptile (homepage)](https://www.greptile.com/): One-page overview + entry point; links to benchmarks, examples, docs, security, and trial.
- [Enterprise](https://www.greptile.com/enterprise): Enterprise evaluation: deployment options, governance/compliance posture, and rollout guidance.
- [Podium case study](https://www.greptile.com/customers/podium): Additional case study focused on high-volume review workflows.
- [Blog](https://www.greptile.com/blog): Blog index (product releases, engineering deep dives, company updates).
- [Content library](https://www.greptile.com/content-library): Index of buyer guides and comparisons (link specific articles when citing).
- [What is AI code review?](https://www.greptile.com/what-is-ai-code-review): Long-form guide explaining AI code review, outputs, and evaluation criteria.
- [Greptile vs Bugbot](https://www.greptile.com/greptile-vs-bugbot): Comparison page: workflow/architecture differences plus benchmark-backed examples.
- [Greptile vs CodeRabbit](https://www.greptile.com/greptile-vs-coderabbit): Comparison page: workflow/architecture differences plus benchmark-backed examples.