faq.mdx

title	Frequently asked questions
description	Answers about Autter reviews, integrations, AI authorship, pricing, and data handling.

The Autter platform reviews pull requests and provides team workflows, integrations, and analytics. The open source Autter CLI records explicit, line-level AI authorship in Git. You can use either product independently or connect the CLI to your Autter organization. No. The Autter CLI uses explicit attribution reported by supported coding agents. It records which lines an agent changed and links those lines to the agent, model, and prompt context. Autter calls this AI authorship and line-level attribution. Yes. Local-only mode works without an account. Prompts remain on the developer machine, while authorship metadata is written to local Git notes. Error and exception telemetry is enabled by default and can be disabled. See [Data and privacy](/cli/data-and-privacy). The platform reviews proposed changes for bugs, edge cases, security risks, codebase-specific patterns, and team rules. It can also use linters, scanners, linked issues, and documentation as context. Yes. Define team standards in natural language and apply them across repositories. Start with advisory feedback, evaluate the findings, and tighten enforcement only after the rule is reliable. See [Custom rules](/guides/custom-rules). The public product supports GitHub and GitLab for code review workflows. The Autter CLI also reads Git repository history independently of the hosted review integration. Autter publicly lists Jira, Linear, Slack, and VS Code alongside GitHub and GitLab. Availability can vary by plan. See [Integrations](/features/integrations) and [current pricing](https://www.autter.dev/pricing). Yes. Autter provides an MCP server so agents such as Claude Code and Cursor can read review findings, reply to comments, re-run reviews, and query the codebase using a scoped access token. See [MCP server](/features/mcp) and [Connect an agent over MCP](/guides/mcp-setup). Data handling depends on the product and mode you choose. In CLI local-only mode, code, prompts, and agent usage data are not uploaded to Autter, although error telemetry is enabled by default. Team deployments process pull request diffs to compute AI percentage without storing those diffs. See [Data and privacy](/cli/data-and-privacy) for the full breakdown. Plans and included pull request volumes are published on the [Autter pricing page](https://www.autter.dev/pricing). Check that page for current limits and features instead of relying on a static value in the docs. No. Autter handles repeatable checks and surfaces context before merge. Human reviewers still own architecture, product behavior, operational tradeoffs, and the final approval decision. If your question is not covered here, use the [Autter contact form](https://www.autter.dev/contact).