# Autter

> Autter is your AI teammate for code review and testing: it reviews code, tests product impact, checks security, governs releases, and closes the loop from production failure to verified fix.

Every page on this site has a markdown twin: append `.md` to the path (the homepage is `/index.md`). Requests sent with an `Accept: text/markdown` header to any page URL also receive markdown. The complete site content in one file is at `/llms-full.txt`.

## Key facts

- **Product**: your AI teammate for code review and testing. Reviews code, tests product impact, checks security, and governs releases: every pull request runs in an isolated sandbox, dependencies are verified against real registries, authorship is classified human vs AI, and the merge is blocked until the checklist clears. Production failures loop back into verified fix PRs.
- **Pricing**: Harbour free (10 PRs/month, value models, reviews pause), Port $20/month (40 PRs, then $1.20/PR, core premium catalog), Cargo $40/month (100 PRs, then $0.90/PR, full catalog), Fleet $80/month (300 PRs, then $0.60/PR, full catalog), Enterprise custom with the full catalog. Full details: https://autter.dev/pricing.md.
- **Company**: founded 2026 by Sagnik (co-founder, tech) and Tanvi (co-founder, GTM); pre-launch with a waitlist; remote (IST).
- **Deployment**: native GitHub integration plus the [Autter CLI](https://autter.dev/features/cli.md); docs at https://autter.dev/docs; code at https://github.com/Autter-dev/.
- **Contact**: support@autter.dev (support) · hello@autter.dev (general) · pr@autter.dev (press) · book the founders at https://cal.com/sagnik-autter/30min.

## Pages
- [Home](https://autter.dev/index.md): Autter is your AI teammate for code review and testing. It reviews every pull request, tests product impact, checks security, governs releases, and closes the loop from production failure to verified fix.
- [How It Works](https://autter.dev/how-it-works.md): How Autter's AI code review and testing works, in twelve steps: connect every signal, build product memory, write the docs, map missions, generate tests, run the sandbox, vet dependencies, classify human vs AI code, trace errors, raise verified fixes, measure the team, and keep watching.
- [About](https://autter.dev/about.md): Autter is the AI teammate for code review and testing that your team has been missing. The Friday deploy that started it, the merge gate we built in response, and the small crew behind it.
- [Blog](https://autter.dev/blog.md): Notes on AI code review, automated testing, release governance, and shipping with AI, written by the team building Autter.
- [Careers](https://autter.dev/careers.md): Join the team building the AI teammate for code review and testing. See open roles at Autter.
- [Changelog](https://autter.dev/changelog.md): Everything Autter ships, the day it ships: the running record of new AI code review and testing features across the platform, CLI, and review engine.
- [Customers](https://autter.dev/customers.md): Autter's AI code review and testing is built for the people who own what ships: engineering leads, open source maintainers, security engineers, and startup CTOs who answer for what happens when code reaches production.
- [Security](https://autter.dev/security.md): How Autter protects the code it reviews: ephemeral sandboxes destroyed after every run, no training on your code, least-privilege access, scoped expiring tokens, a full audit trail, and on-prem deployment for enterprises that need the review plane inside their own network.
- [Pricing](https://autter.dev/pricing.md): Autter pricing is a monthly price for a set number of pull requests. Harbour is free for 10 pull requests a month on value models. Paid plans start at $20, then bill per extra pull request. Premium models start on Port.
- [Contact](https://autter.dev/contact.md): Get in touch with the Autter sales and support teams for AI code review demos, onboarding support, or product questions, or book a call directly with the founders.

## Features
- [Code Review](https://autter.dev/features/code-review.md): Autter's AI code review runs every pull request instead of just reading it: sandbox execution, dependency verification against real registries, and findings that block the merge button until they clear.
- [Pipeline Automation](https://autter.dev/features/pipeline.md): Autter's AI testing pipeline maps critical flows into missions, generates the tests to cover them, runs everything on every pull request, and turns confirmed failures into verified fix PRs.
- [Analytics](https://autter.dev/features/analytics.md): Autter's code review analytics roll quality trends, hotspots, flaky rate, cycle time, and test coverage into a single release-readiness verdict, with every production error traced to the code that caused it.
- [AI Chat](https://autter.dev/features/chat.md): Ask Autter's AI teammate anything about your codebase. Chat is grounded in product memory, so every answer cites the code, tests, owners, and history it came from.
- [Runtime Observability](https://autter.dev/features/runtime.md): Autter Runtime watches production through a lightweight open source SDK, folds raw errors into incidents, traces each one to the deploy and pull request that caused it, investigates against your indexed codebase, and opens the fix as a PR for a human to review.
- [Integrations](https://autter.dev/features/integrations.md): Autter connects GitHub, CI, issue trackers, docs, Sentry, PostHog, Grafana, and deployments into one AI code review context, installed through a native GitHub App in minutes.
- [Autter CLI](https://autter.dev/features/cli.md): The Autter CLI is an open source Git extension for line-level AI code attribution. Know which AI wrote each line, straight from Git history, with autter blame and autter stats.
- [MCP Server](https://autter.dev/features/mcp.md): The Autter MCP server gives Claude Code, Cursor, OpenCode, and any MCP-compatible coding agent direct access to Autter's review findings, codebase intelligence, security scans, and team conventions, so the agent that writes the code can fix its own review.

## Blog posts
- [The PR Queue Wasn’t Built for This Much Code](https://autter.dev/blog/the-pr-queue-wasnt-built-for-this-much-code.md): AI made code abundant. Human attention did not scale with it. Why we built Autter Triage around the changes that actually deserve deeper review. (2026-10-10)
- [How to Prioritize Vulnerabilities: CVSS, EPSS, KEV, and Reachability](https://autter.dev/blog/how-to-prioritize-vulnerabilities.md): How to prioritize vulnerabilities: use CVSS for severity, EPSS for likelihood, KEV for proof, and reachability for reality. Then stop patching in alphabetical order. (2026-10-09)
- [Claude Code, Cursor, Codex: Fast Hands, Zero Fear of Consequences](https://autter.dev/blog/ai-coding-assistant-security-risks.md): The real security risks of Claude Code, Cursor, and Codex: insecure code, leaked secrets, bad packages, and prompt injection. And why a merge gate beats a polite suggestion. (2026-10-05)
- [GitHub Refusing Merge Is Not an SDLC](https://autter.dev/blog/github-refusing-merge-is-not-an-sdlc.md): Why GitHub's real merge levers still stop at the PR — and how Autter extends enforcement from authorship through runtime and repair. (2026-09-12)
- [How RelayCore cut AI coding spend by 33% with Autter](https://autter.dev/blog/relaycore-cut-ai-coding-spend-33-percent.md): A 38-person engineering team used Autter to connect AI costs to repositories, agents, and completed work. Six weeks later, RelayCore had cut monthly AI spend from $31,800 to $21,400 while shipping 22% more AI-assisted pull requests. (2026-09-08)
- [How Northstar cut PR review time by 62% with Autter](https://autter.dev/blog/northstar-cut-pr-review-time-62-percent.md): A 24-person team was generating code faster than it could review it. Autter put a merge gate between each PR and main — median review time fell from 4.2 hours to 1.6 hours. (2026-09-04)
- [Copilot Can Review Copilot. That Is Not a Second Reviewer.](https://autter.dev/blog/copilot-can-review-copilot-that-is-not-a-second-reviewer.md): GitHub now lets Copilot code review cover Copilot-authored PRs. Covering bot PRs was the right product move. Same-vendor review is still the author certifying the merge. (2026-08-30)
- [Your PR Is Green. Should You Still Be Nervous?](https://autter.dev/blog/merge-room-episode-1-your-pr-is-green.md): Tests are green, the code compiles, and the PR looks harmless. Learn what code review must catch beyond CI and why reviewing a diff is only the beginning. (2026-08-21)
- [Autter vs Claude Code Review: Should the AI That Writes Your Code Also Review It?](https://autter.dev/blog/autter-vs-claude-code-review.md): Autter is the stronger first choice for teams that need risk-based verification, enforceable merge gates and predictable code assurance. See how it compares with Claude Code Review. (2026-07-28)
- [Introducing the Autter MCP server](https://autter.dev/blog/autter-mcp-server.md): Connect Claude Code, Cursor, OpenCode, and other MCP-compatible coding agents to Autter’s reviews, codebase intelligence, security findings, and team knowledge. (2026-07-27)
- [Autter vs CodeRabbit: a reviewer is not a merge gate](https://autter.dev/blog/autter-vs-coderabbit.md): Compare Autter and CodeRabbit on AI code review, runtime verification, merge gates, rate limits, and pricing for teams shipping AI-generated code. (2026-07-16)
- [Autter got promoted: From code reviewer to engineering teammate](https://autter.dev/blog/autter-got-promoted.md): Autter started as a code reviewer. Now it stays with a change from review and testing through merge, production, and the verified fix. (2026-07-13)
- [We Open-Sourced the Tool We Use to Track AI-Generated Code](https://autter.dev/blog/we-open-sourced-the-tool-we-use-to-track-ai-generated-code.md): Git says who committed a line, not who actually wrote it. We built Autter to track AI-generated code at the line level, and today the CLI is open source. (2026-06-22)
- [What We Did When Our Email Stack Stopped Sending](https://autter.dev/blog/what-we-did-when-our-email-stack-stopped-sending.md): Our dashboards said delivered. Inboxes said nothing. Two months of quiet outbound later, we built Hedwig and email-checker and open sourced both. (2026-05-23)
- [Why We Block the Merge Button Instead of Posting a Comment](https://autter.dev/blog/why-we-block-the-merge-button-instead-of-posting-a-comment.md): Every code review tool in the market posts comments. We block the merge. That one design decision changed how the entire product got built, what we sell, who we sell it to, and what the moat looks like. (2026-05-14)
- [AI Reviewing Its Own Code Is a Conflict of Interest](https://autter.dev/blog/ai-reviewing-its-own-code-is-a-conflict-of-interest.md): Every AI code generation company is shipping a review product. The author and the reviewer are the same engine with a different prompt. That is not a code review. (2026-05-05)
- [Day 1 of a Startup: A Product, No Users, and Infinite Anxiety](https://autter.dev/blog/day-1-of-a-startup.md): Launching a startup feels less like firing a rocket and more like pushing a paper boat into a river and then sprinting alongside the bank hoping it doesn't sink. (2026-04-28)
- [We scanned 10 startup codebases. Here's what we found.](https://autter.dev/blog/we-scanned-10-startup-codebases.md): Over the past few weeks, we ran manual codebase scans for 10 early-stage startups. Nine out of ten had at least one high-severity finding. (2026-04-10)
- [We Spent 3 Weeks Fighting About Names](https://autter.dev/blog/spent-3-weeks-fighting-names.md): Naming a startup is probably not the most important thing you'll do. But get it wrong and you'll pay for it quietly, in every intro, every cold email, every 'wait, how do you spell that?' on a call. (2026-04-08)
- [Catching AI-Generated Code Before It Ships](https://autter.dev/blog/ai-generated-code-review.md): How engineering teams use autter to detect subtle issues in AI-authored pull requests that pass CI but fail in production. (2026-03-28)
- [Scaling Code Review Across Distributed Engineering Teams](https://autter.dev/blog/scaling-code-review-distributed-teams.md): How autter eliminates timezone bottlenecks and ensures consistent review quality across distributed teams working in different time zones. (2026-03-25)
- [Securing Your CI/CD Pipeline Against AI-Introduced Vulnerabilities](https://autter.dev/blog/automated-pipeline-security.md): How autter's merge gate catches security vulnerabilities that AI coding assistants introduce — before they reach your deployment pipeline. (2026-03-20)
- [We were called Shipgate.](https://autter.dev/blog/we-were-called-shipgate.md): Shipgate made sense on paper, but it never made people feel anything. Then I drew an otter at a harbour gate, and the product (and the name) snapped into focus. (2026-03-19)
- [Enforcing Architecture Decisions at the Merge Gate](https://autter.dev/blog/enforcing-architecture-decisions.md): How autter automatically enforces architectural boundaries, dependency rules, and design patterns — so your ADRs don't become wishful thinking. (2026-03-18)
- [Measuring Code Quality in the Age of AI-Assisted Development](https://autter.dev/blog/team-code-quality-metrics.md): Use autter's analytics to track how AI-generated code impacts your team's quality metrics, review velocity, and production stability. (2026-03-15)
- [The Harbour Is Open](https://autter.dev/blog/the-harbour-is-open.md): AI now writes a meaningful share of your pull requests. The code looks right. The tests pass. And yet something is off in ways that only show up in production. autter is a merge gate built for this new reality. (2026-03-13)
- [Reducing Code Review Fatigue for Senior Engineers](https://autter.dev/blog/reducing-review-fatigue.md): How autter automates the repetitive parts of code review so your senior engineers can focus on architecture, mentoring, and the decisions that actually require human judgement. (2026-03-12)
- [Accelerating New Developer Onboarding with Automated Review](https://autter.dev/blog/onboarding-new-developers.md): How autter helps new team members learn your codebase conventions, patterns, and architecture through real-time feedback on every pull request. (2026-03-10)
- [Building Compliance-Ready Audit Trails for Every Code Change](https://autter.dev/blog/compliance-audit-trails.md): How autter provides automated, tamper-evident audit trails for every pull request — meeting SOC 2, ISO 27001, and HIPAA requirements for change management. (2026-03-08)
- [Managing Code Review at Scale in Monorepos](https://autter.dev/blog/monorepo-code-review.md): How autter handles the unique challenges of monorepo code review — cross-package impact analysis, ownership routing, and consistent standards across hundreds of packages. (2026-03-05)

## Open roles
- [Frontend Engineer](https://autter.dev/careers/frontend-engineer.md): Work directly on the Autter web application, shipping real features alongside the founding team.
- [Backend Engineer (Agentic AI)](https://autter.dev/careers/backend-engineer.md): Build and maintain the systems that detect blast radius, security vulnerabilities, and AI-generated code quality issues.
- [Marketing Intern](https://autter.dev/careers/marketing-intern.md): Own growth, GTM, content, campaigns, creative, and community work with the founding team before launch.

## Optional
- [Terms of Service](https://autter.dev/terms-of-service.md): Autter Terms of Service: the rules and guidelines governing your use of the Autter AI code review and testing platform, CLI, and related services.
- [Privacy Policy](https://autter.dev/privacy.md): Autter Privacy Policy: how we collect, use, and protect your information when you use the Autter AI code review and testing platform, CLI, and related services.
- [Full site content](https://autter.dev/llms-full.txt): every page of this site as one markdown file