Menu
ContextBridge
How it works Works with Pricing FAQ
EnglishTiếng Việt简体中文日本語
Open the app

Terms of Service

Last updated: 2026-10-08

The agreement between you and the operator of ContextBridge. Short version: the tool is free, the output is yours, you are responsible for what you do with it, and there is no warranty.

This document was prepared by the operator of this site and is not legal advice. If you are relying on it for a compliance decision, have a qualified lawyer in your jurisdiction review it.

On this page

  1. 1. Agreement
  2. 2. What the Service is
  3. 3. Licence to use
  4. 4. Your code and the output
  5. 5. Your responsibility for what you send
  6. 6. No warranty about the output
  7. 7. Fair use and rate limits
  8. 8. Advertising and funding
  9. 9. Third-party services
  10. 10. Limitation of liability
  11. 11. Indemnity
  12. 12. Suspension and termination
  13. 13. Changes to the Service and to these Terms
  14. 14. Governing law and disputes
  15. 15. General

1. Agreement

These Terms of Service ("Terms") govern your use of the ContextBridge website and client-side software (the "Service"), operated by ContextBridge ("we", "us"). By accessing or using the Service you agree to these Terms. If you do not agree, do not use the Service.

You must be old enough to enter a binding contract in your jurisdiction, or have a parent or guardian's permission. The Service is not directed at children under 13, and we do not knowingly collect data from anyone, of any age.

2. What the Service is

The Service is a static web application that runs entirely in your browser. It parses source code you select, using tree-sitter compiled to WebAssembly, and generates Markdown documents describing the structure of that code.

  • There is no account, no registration and no sign-in.
  • There is no server-side storage of your files, your code, or the documents you generate.
  • There is no upload path. The Service contains no endpoint that accepts your source code.
  • The generated documents exist only in your browser's memory until you choose to download, copy or share them.
  • This is a factual description of the architecture, and it is the basis of the privacy representations in our Privacy Policy. If a future version changes it, these Terms and that Policy change first.

    • There is no account, no registration and no sign-in.
    • There is no server-side storage of your files, your code, or the documents you generate.
    • There is no upload path. The Service contains no endpoint that accepts your source code.
    • The generated documents exist only in your browser's memory until you choose to download, copy or share them.

    3. Licence to use

    We grant you a personal, non-exclusive, non-transferable, revocable licence to use the Service for lawful purposes, subject to these Terms and the Acceptable Use Policy.

    The Service itself is open-source software licensed under the MIT Licence. Portions are derived from Understand-Anything (also MIT), attributed in the application's documentation. Nothing in these Terms restricts your rights under those licences.

    4. Your code and the output

    You retain all rights to the code you analyse. We do not acquire any licence, claim or interest in it, and we never receive it.

    The generated documents are yours. They are derived from your own source code, and you may use, modify, publish, sell and distribute them for any lawful purpose, including commercially, with no attribution requirement and no restriction.

    We do ask — without requiring — that you do not represent generated documents as human-written analysis, since they contain no human judgement.

    5. Your responsibility for what you send

    The Service generates documents from your code so that you can provide them to a third-party AI system of your choosing. That decision is yours alone, and it is consequential.

    A generated document can contain file paths, symbol names, and — where you enable that option — verbatim source code, which may include string literals containing credentials, connection strings, internal hostnames or personal data.

  • You are responsible for reviewing what you send, and for complying with your own obligations to your employer, your clients and applicable law.
  • Third-party AI services have their own terms. Once you paste or upload a document into one, your use of it is governed by that provider's agreement with you, not by ours. We have no relationship with those providers and no ability to retrieve or delete what you send them.
  • Do not use the Service to prepare content for transmission to a third party where doing so would breach a confidentiality obligation you owe.
  • A secret-redaction feature is on our published roadmap. Until it ships, treat every generated document as confidential to the same degree as the source code it describes.

    • You are responsible for reviewing what you send, and for complying with your own obligations to your employer, your clients and applicable law.
    • Third-party AI services have their own terms. Once you paste or upload a document into one, your use of it is governed by that provider's agreement with you, not by ours. We have no relationship with those providers and no ability to retrieve or delete what you send them.
    • Do not use the Service to prepare content for transmission to a third party where doing so would breach a confidentiality obligation you owe.

    6. No warranty about the output

    The documents are generated by deterministic static analysis, and we take care that they are accurate about what they do contain. But you must understand precisely what that means.

  • Line ranges, symbol names, parameter lists and import relationships are extracted from a parse tree of the code as it existed at generation time. They are accurate, or the file has since changed.
  • The documents contain no descriptions of behaviour. They do not tell you what your code does, and they are not a substitute for reading it, for tests, or for review by a qualified engineer.
  • Coverage may be partial. Files in languages without a loaded grammar are listed but not parsed. Where this happens the document says so explicitly, in a warning block, and names the affected files. You must read that warning.
  • Analysis of code the Service cannot parse, of obfuscated code, of generated code, or of files above the size limit may be incomplete or absent.
  • Any conclusion an AI system draws from a generated document is that system's, not ours. We do not warrant it, and we are not responsible for it.

    • Line ranges, symbol names, parameter lists and import relationships are extracted from a parse tree of the code as it existed at generation time. They are accurate, or the file has since changed.
    • The documents contain no descriptions of behaviour. They do not tell you what your code does, and they are not a substitute for reading it, for tests, or for review by a qualified engineer.
    • Coverage may be partial. Files in languages without a loaded grammar are listed but not parsed. Where this happens the document says so explicitly, in a warning block, and names the affected files. You must read that warning.
    • Analysis of code the Service cannot parse, of obfuscated code, of generated code, or of files above the size limit may be incomplete or absent.

    7. Fair use and rate limits

    The Service is free, and keeping it free requires that it not be abused. We apply automated rate limits at the network edge, currently expressed as a maximum number of analyses per hour and per day per source address.

    These limits are set well above ordinary human use — they exist to stop automated scraping, credential-stuffing-style hammering and denial-of-service, not to ration developers. Current values are published in the application's configuration and may be adjusted.

    You agree not to: circumvent or probe the rate limits; use the Service as a component of an automated pipeline that generates load disproportionate to a human user; resell access to the Service; or attempt to interfere with its operation.

    If you need programmatic access at volume, contact us — we would rather discuss it than block you.

    8. Advertising and funding

    Advertising is disabled by default and will only be enabled with your explicit, recorded opt-in. Where advertising is shown, it is disclosed at the point of display, the opt-out is a single action and is persisted, and the Privacy Policy is amended before any advertising script is introduced.

    Voluntary support links (for example to Ko-fi, GitHub Sponsors or Open Collective) are provided by third parties. We receive no data from them beyond the fact of a donation, and we cannot refund, dispute or reverse a payment you make to them.

    Some outbound links may be affiliate links. These are always disclosed as such at the point of display, load no script, set no cookie, and cost you nothing additional.

    9. Third-party services

    The Service is hosted on Cloudflare Pages. Use of that infrastructure is governed by Cloudflare's terms, and Cloudflare acts as our infrastructure provider. See the Privacy Policy for what this means for your IP address.

    The Service links to third-party AI products, editors and agents. Those links are provided as convenience. We do not control those services, do not warrant them, and are not responsible for anything that happens when you use them.

    10. Limitation of liability

    The Service is provided free of charge, "as is" and "as available", without warranty of any kind, whether express, implied or statutory — including without limitation the implied warranties of merchantability, fitness for a particular purpose, and non-infringement.

    To the maximum extent permitted by applicable law, we and our contributors are not liable for any indirect, incidental, special, consequential or punitive damages, or any loss of profits, revenue, data, goodwill or anticipated savings, arising out of or in connection with your use of the Service.

    This includes, without limitation: decisions you or an AI system make based on a generated document; a document that omits part of your codebase; disclosure of confidential information that you chose to send to a third party; unavailability of the Service; or the acts of a third-party AI provider.

    Our total aggregate liability for any claim arising out of or relating to these Terms or the Service shall not exceed the greater of (a) the amount you paid us in the twelve months preceding the claim, which for the free Service is zero, and (b) USD $50.

    Some jurisdictions do not permit the exclusion of implied warranties or the limitation of liability for certain damages. In those jurisdictions, the exclusions and limitations above apply only to the extent permitted by law, and you may have additional rights.

    11. Indemnity

    You agree to indemnify and hold harmless ContextBridge and its operators from claims arising out of (a) your breach of these Terms or the Acceptable Use Policy, (b) your use of a generated document, including providing it to a third-party AI service, and (c) your violation of any law or of a third party's rights.

    We will notify you of a claim for which we seek indemnity, but a failure to notify does not waive your obligation except where you are materially prejudiced by the delay.

    12. Suspension and termination

    We may suspend or block access where required to protect the Service, its users or a third party — for example in response to abuse, an attack, or a legal obligation. Because there are no accounts, blocking is applied at the network level and is normally temporary.

    You may stop using the Service at any time; there is nothing to cancel and no data to delete, because we hold none.

    Sections 4, 5, 6, 10, 11, 13 and 14 survive termination.

    13. Changes to the Service and to these Terms

    The Service is under active development. Features may be added, changed or removed, and language coverage, file limits and rate limits may change.

    We may revise these Terms. Where a revision materially reduces your rights or increases your obligations, we will indicate the change by updating the "Last updated" date at the top of this page and, where practical, by a notice in the application. Continued use after a change constitutes acceptance of the revised Terms.

    If you do not accept a revision, your remedy is to stop using the Service. Because it is free and holds no data of yours, this remedy is complete.

    14. Governing law and disputes

    This clause requires your attention before launch. These Terms are currently drafted to be governed by the laws of the Socialist Republic of Vietnam, without regard to conflict-of-law rules, and the competent courts of Vietnam have exclusive jurisdiction.

    Nothing in these Terms deprives a consumer of the protection of mandatory provisions of the law of the country where they are habitually resident. If you are a consumer in the EU, the UK, or another jurisdiction with mandatory consumer protections, those protections apply to you notwithstanding the choice of law above.

    Before initiating any formal proceeding, both parties agree to attempt in good faith to resolve the dispute by written notice to the contact address below, and to allow 30 days for a response.

    15. General

    These Terms, together with the Privacy Policy, Cookie Policy, Acceptable Use Policy and Disclaimer, constitute the entire agreement between you and us regarding the Service.

    If a provision is held unenforceable, the remainder continues in full force, and the unenforceable provision is replaced by an enforceable one that most closely reflects the original intent.

    Our failure to enforce a provision is not a waiver of it. A waiver is effective only in writing and only for the specific instance.

    You may not assign these Terms without our consent. We may assign them to a successor in connection with a transfer of the Service, with notice.

    Questions, notices and abuse reports: hello@contextbridge.app (general), abuse@contextbridge.app (abuse and legal process).

    ← Back to the app

    ContextBridge

    ContextBridge reads your project with tree-sitter — a real source-code parser, not regex, not guesswork — and turns it into a token-budgeted Markdown map: every symbol, signature, line range and dependency edge. Then you drop that one file into any AI chat and it finally understands your architecture.

    No trackers. No cookies. No analytics.

    Legal

    • Terms of Service
    • Privacy Policy
    • Cookie Policy
    • Disclaimer
    • Acceptable Use Policy
    • Data Processing Information

    Resources

    • Works with
    • FAQ
    • Security
    • Report abuse
    • Contact

    Language

    EnglishTiếng Việt简体中文日本語

    © 2026 ContextBridge (sole proprietorship, Vietnam). All rights reserved.

    The tool is MIT licensed. Portions derived from Understand-Anything (MIT).