
Code Input
Code Input is a web-native productivity suite for development teams. It resolves Git merge conflicts in the browser, manages CODEOWNERS files, queues merges, automates repository workflows and tracks engineering metrics, with optional bring-your-own-key AI suggestions.
What is Code Input?
Code Input is a web-native platform for development teams, built around five products that work together or on their own. Its stated ambition is to be the web layer for high-velocity software delivery, and everything happens in the browser rather than in a terminal or an IDE. Merge Conflicts is the entry point: a three-way editor showing the ancestor, your version and theirs, with side-by-side diffs, syntax highlighting, one-click acceptance of either side, direct editing and a push straight to GitHub. Resolutions are validated to catch syntax errors, and contextual AI suggestions are available. A live demo and a patch sandbox let you try it without an account. CODEOWNERS offers five views, File Ownership, Patterns, Owners, Tags and Reports, with multiple CODEOWNERS files, ownership declared inside files, free-form tags such as ui, api or infrastructure, coverage rates and pattern efficiency scores. It stays fully backward compatible with GitHub syntax, so no migration is needed. Merge Queue sits above GitHub's own queue and adds priorities, dependencies, sequential automated tests, early conflict detection, real-time monitoring and concurrency controls. Workflow Automation, positioned as an alternative to GitHub Actions, triggers on repository events such as a pull request opened, a security alert or a critical file changed, then responds with reviewer assignment, test runs, email or Slack notifications, API calls, pull request comments and issue creation, filtered by branch, author or files, with a full execution history. Insights adds repository analytics, visual dashboards, DORA metrics, pull request tracking, trend analysis and customizable PDF reports. Access runs through the web app, a Chrome and Firefox extension that surfaces CODEOWNERS inside GitHub's own interface, and an MIT-licensed Rust command-line tool installable via npm, Cargo, AUR or platform binaries. Sign-in uses GitHub, GitLab or Bitbucket OAuth, or email, plus a GitHub App installation. The AI layer is bring your own key: you supply an OpenAI, GitHub Copilot or Anthropic Claude key, and Enterprise customers can request a custom AI integration. On security, the publisher states that source code is never stored and that all code processing runs inside your own GitHub Actions runners.
What it does
- Resolve Git merge conflicts in the browser with a three-way editor showing ancestor, our version and their version
- Get AI resolution suggestions using your own OpenAI, GitHub Copilot or Anthropic Claude key
- Push the resolved file straight to GitHub from the web editor
- Manage advanced CODEOWNERS: multiple files, ownership declared inside files, tags and ownership queries
- Queue pull requests for merge with priorities, dependencies and automated tests
- Automate reactions to repository events: email and Slack notifications, reviewer assignment, pull request comments
- Track engineering and DORA metrics on dashboards and export PDF reports
When to use Code Input / When not to
A quick filter to help you decide if Code Input is the right fit.
When to use Code Input
- Solo developers, who get a permanent free tier with unlimited public repositories and unlimited conflict resolutions
- Development teams that hit frequent merge conflicts and would rather settle them in the browser than in the terminal
- Teams working on monorepos or large repositories, thanks to multiple CODEOWNERS files, in-file ownership and tags
- Engineering managers looking for DORA metrics, pull request tracking and exportable dashboards
- GitHub-first organizations, since the GitHub App, direct push and Actions runners make the integration native
When not to use Code Input
- Teams that do not use Git or a hosted repository platform
- GitLab or Bitbucket shops: those providers only appear as sign-in methods, while every documented feature targets GitHub
- Anyone who needs a mobile app, as there is no iOS or Android application
- Integrators who want to drive the tool through an API: API access sits on the Enterprise tier and no API reference is published
- Organizations that require formal compliance commitments, since no DPA, no subprocessor list and no SOC 2 or ISO 27001 certification is published
How to use Code Input
A typical end-to-end flow, from setup to results.
- Try it first without an account: open the live merge conflict demo or upload your own patch to the diff sandbox
- Create an account by signing in with GitHub, GitLab or Bitbucket, or register by email
- Install the GitHub App on your organization, which may require an organization administrator's approval
- Run the initial configuration described in the Getting Started documentation
- Connect the repositories you want to track
- Optional: install the Rust command-line tool via npm, Cargo, AUR or a platform binary
- Optional: add the Chrome or Firefox extension to see CODEOWNERS on GitHub pull request pages
- Open a merge conflict in the web editor, compare side by side, accept one side or edit directly, then push to GitHub
- Optional: plug in your own OpenAI, GitHub Copilot or Anthropic Claude key to get AI resolution suggestions
- Define your CODEOWNERS rules, check coverage in the Reports view, then build workflows and follow executions and Insights metrics
Pros & Cons
Pros
- A genuinely usable permanent free tier: one user, unlimited public repositories, unlimited conflict resolutions, five workflows and 10,000 invocations
- A low paid entry point at 9 USD per month, which unlocks unlimited private repositories
- Nothing to install for the web app, setup announced at two minutes and no credit card required
- Source code is never stored, since processing happens inside the customer's own GitHub Actions runners
- Bring-your-own-key AI, so you keep control of your provider and your token budget
- Open source command-line tool and browser extension, with full backward compatibility with GitHub CODEOWNERS syntax and zero migration
- Public, readable pricing with a detailed comparison table across the four tiers
Cons
- Heavy dependence on GitHub: the GitHub App, direct push, merge queue and Actions runners are built for GitHub, while GitLab and Bitbucket appear only as sign-in methods
- No published compliance commitment: no data processing agreement, no subprocessor list, no SOC 2 or ISO 27001 certification, and no explicit mention of the GDPR in the policy text
- Data hosting is not localized: certified data centres are mentioned, but no country and no region
- No retention period is quantified anywhere in the privacy policy
- API access is reserved for the Enterprise tier and no API documentation is published
- AI is not included: you must bring your own key, which means an account and a token budget with a third party
- A young and very small publisher with a blurred legal identity, the terms pointing to Malaysia while the structured data gives a San Francisco address
Pricing & Plans
A permanent free plan is available at 0 USD per month. The lowest paid entry point is the Pro plan at 9 USD per month, followed by Team at 19 USD per user per month; Enterprise pricing is available on quotation. A free trial is offered without a credit card. The pricing page carries a billing period selector, but no annual figure is displayed.
- 1 user
- unlimited public repositories
- unlimited conflict resolutions
- 5 workflows
- GitHub and email integration
- 10
- 000 invocations
- three-way editor
- 1 user
- unlimited private repositories
- unlimited workflows
- Slack integration
- advanced CODEOWNERS interface
- Chrome and Firefox extension
- DORA metrics
- email and Slack notifications
- everything in Pro plus unlimited seats
- advanced CODEOWNERS
- merge queue with concurrency controls and monitoring
- collaboration tools
- pull request tracking
- PDF reports
- priority email and voice support
- 1 million invocations per seat then 3 USD per additional million
- GitHub Enterprise support
- custom billing
- on-premise deployment
- 24/7 support
- SSO and SAML
- API access
- custom AI integration and dedicated support
Data, GDPR & hosting
A consolidated view of how Code Input handles your data.
GDPR overview
Implementation is partial and never declared. The visible text of the terms and the privacy policy never uses the words GDPR or General Data Protection Regulation; the only trace sits in JSON-LD metadata, where the terms describe a Data Protection section as covering GDPR compliance. That is machine-readable metadata, not a written commitment. The privacy policy does grant the usual set of rights: access, correction, deletion, objection to processing, portability and withdrawal of consent. Beyond that, nothing. No legal basis is cited, no data protection officer is named, no Article 27 EU representative is designated, no data processing agreement is offered, no subprocessor list exists and no international transfer mechanism is documented. There is no legal or privacy contact address, only contact, support, sales, enterprise and founder mailboxes. The terms name Malaysia, outside the EU, as the governing country.
Who owns the data?
The terms state that the Service and its original content, features and functionality remain the exclusive property of the company and its licensors. In the other direction, no clause grants the publisher any licence over your code or your content, and the publisher says source code is never stored: processing runs inside your own GitHub Actions runners. The privacy policy therefore covers personal data only, namely name and contact details, account identifiers, payment information, preferences, usage data, device information, logs and performance metrics. Sharing is limited to four listed cases: service providers, advisers and auditors, authorities acting under legal process, and partners with your consent. No subprocessor list and no data processing agreement are published.
Reuse rights
Your own material stays yours: nothing in the terms grants Code Input a licence over your source code or content, and the publisher states that code is never stored. The Service's own content is closed, however. You may not copy, modify, distribute, sell, lease or otherwise exploit any content or materials made available through the Service without the company's prior written consent. Two components are an explicit exception: the Rust command-line tool is published under the MIT licence and the browser extension is announced as fully open source, so both can be reused and redistributed freely under those terms. On personal data, the privacy policy grants access, correction, deletion, objection to processing, portability and withdrawal of consent, and nothing indicates that customer data is used to train models.
Data retention & training
Hosting summary
The privacy policy says only that information is stored on secure servers located in certified data centres, protected by industry-standard security measures. No country, no region and no hosting provider is named, and there is no mention of international transfers, standard contractual clauses or any European location. Code processing is a separate matter: according to the publisher, source code is never stored and all processing runs inside the customer's own GitHub Actions runners, outside Code Input's infrastructure. On jurisdiction, the picture is contradictory: the terms of service state that the governing country is Malaysia, while the structured data published on the site gives a San Francisco, California address, and nothing on the site reconciles the two. At network level, the domain resolves behind Cloudflare on an anycast address, which says something about the delivery infrastructure but nothing about where personal data is actually held. In short, hosting is described in general terms only, and anyone with a data residency requirement will have to ask the publisher directly.
Things to keep in mind
Risks and trade-offs to weigh before adopting Code Input.
- AI merge suggestions can be plausible and wrong: a resolution accepted without review can introduce a silent regression that no test catches
- Skill erosion: resolving conflicts in one click can slowly cost you the understanding of Git history and of the underlying commands
- The AI runs on your own key, so the code you send is governed by the terms of that third-party provider, not by Code Input's
- Sign-in uses GitHub OAuth with repository permissions: check the scope granted, because a compromised account exposes every connected repository
- Team metrics and individual scores covering productivity, quality, collaboration and velocity invite managerial misuse and the perverse effects of individual performance measurement
- No data processing agreement, no subprocessor list and no named hosting country: the processing chain cannot be documented for an audit
- No quantified retention period and a very small publisher whose durability is unproven, with an ambiguous legal identity and liability capped at 100 USD: plan an exit route
Setup & Integrations
Technical difficulty
Low for the intended audience. The prerequisites, a GitHub account and the right to install a GitHub App, are familiar to developers; setup is announced at two minutes and nothing has to be installed locally for the web app. Four Getting Started pages cover account creation, GitHub App installation, configuration and the command-line tool. Two caveats: installing the App on an organization may need an administrator's approval, and the optional steps, the command-line tool and your own AI key, are more technical. Advanced CODEOWNERS and the merge queue assume you understand ownership rules and your existing CI.
Deployment
Integrations
Supported languages
Behind Code Input
Social
Resources
All the official URLs gathered for verification and reference.
Alternatives
Tools that compete with or complement Code Input.
Frequently asked questions
What is Code Input?
Do I need to install anything?
How does merge conflict resolution work?
Which AI providers are supported?
Is my source code secure?
Can I try Code Input for free?
How much does it cost?
Do I have to migrate my existing CODEOWNERS file?
Is there an API?
Where is my data hosted?
Should you pick Code Input?
Code Input is a narrowly targeted tool, and that is its strength: it attacks a precise, daily pain point of team work with Git, and it does so without asking anyone to leave the browser. The value for money is favourable, with a free tier that is genuinely usable and a first paid step at 9 USD a month, which puts the product within reach of a solo developer as easily as of a small team. Its dependence on GitHub is assumed rather than hidden. The GitHub App, the direct push, the merge queue and the use of your own Actions runners are what make the experience smooth, and they are also what makes the product hard to recommend to a GitLab or Bitbucket shop. The AI layer is real but secondary and not included: suggestions run on your own OpenAI, Copilot or Anthropic key, which keeps you in control of your provider and your budget, but also means the AI is not part of what you buy. The point to weigh before a company purchase is the publisher itself. Code Input is young and very small, with a single visible founder and two open positions, and its legal transparency is thin: the terms name Malaysia, the structured data on the site gives a San Francisco address, liability is capped at 100 USD and no incorporation details are published. Compliance is not formalised either, with no data processing agreement, no subprocessor list, no certification and no named hosting country, which will rule the tool out for regulated organisations. Two things soften the risk: source code is never stored, and the command-line tool and browser extension are open source, so the CODEOWNERS side of the workflow does not lock you in.
- Choosing a selection results in a full page refresh.
- Opens in a new window.