Privacy Policy
Effective date: September 18th, 2026
Last updated: October 3rd, 2026
1. Who we are
The Capitol Ledger ("Capitol Ledger", "we", "us") is an independent civic transparency project operated by Kartik Tyagi at capitolledger.com.
For any privacy question or request, contact: support@capitolledger.com
We are a small independent project. We do not sell personal data, we do not run advertising, and we do not use your data to build advertising or marketing profiles.
2. The short version
You can read Capitol Ledger's published articles without creating an account. Our hosting providers process technical request data, and we use aggregate analytics as described below.
If you choose to sign in with Google, we create an account and store your name, email address, and profile picture to run it. You can delete your account at any time from your Profile page.
If you use Ledger AI, your questions and recent conversation are processed by our backend and Google's Gemini API to find relevant material and generate answers. We keep account-linked usage records, but do not save conversation transcripts in our database. Your browser keeps the current chat for the tab's session. Sections 3.2 and 5 explain this processing, including Google's separate retention.
We also receive information you choose to send through our feedback form or by email.
3. Information we collect
3.1 Accounts
Signing in is optional. If you choose to sign in with Google, we create an account for you and store:
- your name, email address, and profile picture, as provided by Google
- when your account was created and last updated
- Ledger AI usage counters: how many questions you've used in the current 24-hour period, and when that count resets
We get this information directly from Google when you sign in, through Supabase Auth, our authentication provider — we never ask you to create a separate password. We use it to operate your account and keep you signed in.
You can delete your account at any time from your Profile page. Deleting your account removes your login, profile record, and account-linked Ledger AI usage records. Successful deletion also clears the saved Ledger AI chat in the browser tab used to delete the account. Other browser copies and providers' technical logs have separate retention, described below.
3.2 Ledger AI questions and usage
Ledger AI requires sign-in. When you ask a question, we process its text, recent messages from the conversation, and any bill or record you selected. We send the question and relevant conversation context to Google's Gemini API to find relevant documents and generate an answer, together with retrieved public material, saved AI-generated analysis, and results from our document-search tools. Personal information you include in a question is part of that processing, so please do not submit sensitive or confidential information or other people's private information.
We do not save conversation transcripts in our database. We store usage records linked to your account to enforce limits, understand demand and processing costs, and investigate reliability or abuse. These records include:
- the time and outcome of a request, such as an answer, error, or limit being reached
- the lengths of questions, conversation history, and answers, plus message and token counts
- the names and number of research tools used, the number of sources retrieved, and response timings
- whether the chat focused on a bill, a Congressional Record, or general research
These usage records contain no question or answer text, tool-input text, or list of individual source documents. They are deleted when you delete your account. Diagnostic logs are separate and are described in section 3.5.
Your browser stores the current conversation and its source links in session storage so refreshing the page does not lose the chat. This is a local copy, not a saved chat history in your account. Choosing New chat clears it. Successful account deletion also clears it in the tab used to delete the account. Signing out alone does not clear it, and deletion does not remotely erase copies in other tabs or devices; on a shared device, clear the chat or the site's browser data before leaving. Session storage normally lasts for the tab's session, though your browser may restore it when restoring a session.
Google's processing and retention of API content are described in section 5.
3.3 Feedback form submissions
Our feedback form is provided by Formspree, a third-party form service. When you submit the form, we collect:
- your name
- your email address
- the content of your message
Formspree also receives technical metadata alongside your submission, which may include your IP address and browser information, and uses it for delivery and abuse prevention. This is Formspree's processing under their own privacy policy, which you can read at https://formspree.io/legal/privacy-policy/. We do not use a CAPTCHA on this form, so no CAPTCHA provider receives your data.
Your submission is stored on Formspree's servers and forwarded to us. We use it to read and act on your feedback, to correct errors you report, and to reply to you. We do not add you to any mailing list, we do not use your address for marketing, and we do not share it with anyone.
Please don't include sensitive personal information in the form. For anything concerning your personal data, contact us at support@capitolledger.com instead. That reaches us directly and does not pass through Formspree.
3.4 Analytics
We use Vercel Web Analytics to understand how the site is used. It records:
- pages viewed
- the country the request came from
- the referring source (for example, whether you arrived from Google, Facebook, or a direct link)
- general device and browser type
Vercel Web Analytics is designed to be privacy-preserving. It does not set tracking cookies and does not create a persistent identifier that follows you across sites or sessions. We see aggregate patterns (how many people read a given article, roughly where in the world they are, how they found us) rather than individual browsing histories. We cannot use it to identify you.
We also use Vercel Speed Insights to measure how quickly pages load. It records page performance measurements (such as load and rendering times) along with the page visited and general device, browser, and connection type. Like Web Analytics, it is cookieless, does not identify you, and is reported to us only in aggregate.
3.5 Server logs
Our hosting and backend providers automatically log technical request data, including IP addresses, timestamps, and user agent strings. Our backend also logs authentication and rate-limit events, which can include your account identifier, along with operational diagnostics such as error types, counts, and timings. Ledger AI application logs exclude questions, conversation text, answers, free-text tool inputs, and raw provider error messages or tracebacks. These logs are separate from the usage records described above and exist for security, abuse prevention, and debugging. They follow our providers' log-retention settings and are not individually removed when you delete your account. We do not use them for advertising or marketing profiles. Google's separate API content retention is described in section 5.
3.6 Cookies and browser storage
We do not use advertising, marketing, or cross-site tracking cookies, and we do not use cookies to build a profile of you. Our analytics is cookieless.
Our hosting provider may set strictly necessary cookies for security and load balancing. If you sign in with Google, Supabase Auth sets a strictly necessary session cookie (named sb-<project-ref>-auth-token, sometimes split into several numbered cookies) so that you stay signed in. It is set only if you sign in, and we do not use it for tracking or advertising.
Because we use only strictly necessary cookies, we do not display a cookie banner. If we ever add anything beyond that, this policy will be updated and consent obtained first.
If you switch between light and dark mode, the site remembers your choice in your browser's local storage. That setting stays on your device, is never sent to us, and you can clear it at any time through your browser settings.
Ledger AI uses the browser session storage described in section 3.2 to keep the current conversation available after a refresh. Recent messages are sent with your next question so Ledger AI can answer follow-ups.
3.7 What we do not collect
We never see or store a password for your account — sign-in is handled entirely by Google and Supabase Auth. We do not take payments or donations, so we never receive payment information. We do not ask you to provide special category data such as health, biometric, religious, or political affiliation information. If you put such information in a chat question or feedback message, it will be processed as part of that message; please leave it out.
4. Why we process your information, and our legal basis
If you are in the UK, EU, or another region with similar law, these are our purposes and legal bases:
| What we do | Why | Legal basis |
|---|---|---|
| Publish and serve the site | Core service | Legitimate interests |
| Provide and operate your account, if you sign in | You asked us to, by signing in | Performance of a contract / steps taken at your request |
| Provide Ledger AI answers, including sending questions and recent conversation to Gemini | Respond to the questions you choose to ask | Performance of a contract / steps taken at your request |
| Keep account-linked Ledger AI usage records | Manage usage limits, understand demand and costs, and investigate reliability or abuse | Legitimate interests |
| Read and act on feedback submissions | You chose to contact us | Legitimate interests |
| Reply to you, if you gave us an address | You asked for a response | Legitimate interests |
| Aggregate analytics | Understand what's useful and improve the site | Legitimate interests |
| Security, abuse prevention, and rate limiting | Keep the service available and safe | Legitimate interests |
We do not use your data for automated decision-making that produces legal effects for you.
5. AI processing
Capitol Ledger uses Google's Gemini models to analyse public congressional documents and produce the summaries you read on the site. Ledger AI also uses Gemini to process questions, find relevant source material, and generate conversational answers.
We do not automatically include your account name, email, profile picture, sign-in credentials, feedback submissions, IP address, or general browsing history in model prompts. We do include what you type into Ledger AI and relevant recent conversation, along with source material. If you type personal information into a chat, it will therefore be sent to Gemini. Google also receives technical information associated with our API requests under its own terms.
We use the Gemini API through our billing-enabled Google Cloud project under Google's Paid Services terms. Under those terms, prompts and responses are not used to improve Google's products and are processed under Google's data-processing terms.
Google separately retains prompts, context, and responses for 55 days for safety, abuse monitoring, and required disclosures, according to its abuse-monitoring policy. Authorised Google personnel may review flagged content. That policy permits use for models specifically supporting policy enforcement. Deleting your Capitol Ledger account does not itself delete Google's logs.
We do not use Ledger AI conversations to train our own models. For the reliability and coverage limits of AI answers, see our Terms of Service.
6. Who we share information with
We do not sell your personal information and we do not share it for advertising. We disclose it only to the service providers we need to operate, and only for that purpose:
| Provider | Role | Location |
|---|---|---|
| Vercel Inc. | Website hosting and analytics | United States |
| Formspree, Inc. | Feedback form submissions | United States |
| Supabase Inc. | Published congressional content, authentication, account storage, and account-linked Ledger AI usage records | United States |
| Google Cloud Platform | Backend compute for our publishing pipeline and Ledger AI requests | United States |
| Google LLC | Google sign-in (OAuth), Gemini analysis of public documents, and processing Ledger AI questions, recent conversation, and responses | United States; API content may be processed internationally as described below |
We may also disclose information where we are legally required to, in response to a valid legal process, or where necessary to protect our rights, safety, or the integrity of the service. If the project is ever transferred to another operator, this data may transfer with it, and we will post notice before that happens.
7. International transfers
We operate from the United States and our providers are primarily in the United States. If you access the site from outside the United States, your information will be transferred to and processed in the United States, which may not offer the same level of data protection as your home jurisdiction.
Google's Gemini API terms also permit transient storage or caching of API content in countries where Google or its agents operate facilities.
Where required, these transfers rely on the Standard Contractual Clauses or equivalent safeguards in our providers' data processing terms.
8. How long we keep information
| Data | Retention |
|---|---|
| Account data | Until you delete your account |
| Ledger AI usage records in our database | Until you delete your account |
| Ledger AI questions and answers | Processed to answer requests; no conversation transcripts saved in our database. Browser copies and Google's logs have the separate retention below; diagnostic logs follow the server-log row |
| Current chat in browser session storage | For the tab's session, subject to browser session restoration; cleared in the current tab after successful account deletion, or with New chat or the browser's site-data controls |
| Gemini API safety logs held by Google | 55 days under Google's abuse-monitoring policy linked in section 5 |
| Feedback submissions | Held in our Formspree account, subject to Formspree's retention settings; we delete submissions we no longer need and honour deletion requests |
| Email correspondence | 2 Years |
| Analytics | Aggregate only, retained per Vercel's default retention |
| Server logs | Per our providers' log-retention settings; account deletion does not individually erase existing log entries |
You can ask us to delete a feedback submission sooner. See below.
9. Your rights
Depending on where you live, you may have the right to:
- access the personal information we hold about you
- correct inaccurate information
- delete your information
- object to or restrict our processing
- receive a copy of your data in a portable format
- not be discriminated against for exercising these rights
To exercise any of these, email support@capitolledger.com. For account data, you can also act directly: delete your account at any time from your Profile page. We will respond to email requests within 30 days.
In practice, this can include a feedback submission or email you sent us, account information, account-linked Ledger AI usage records, and technical logs. If you want a feedback submission or email deleted, tell us roughly when you sent it and what it was about, and we will remove it from our Formspree account and our inbox. You can delete your account and its Ledger AI usage records from your Profile page; successful deletion also clears the saved chat in that browser tab. Clear copies in other tabs or devices separately using New chat or your browser's site-data settings. We cannot provide a stored conversation history from our database because we do not keep one.
California residents: we do not sell or share personal information as those terms are defined under the CCPA/CPRA, and we do not process sensitive personal information for the purpose of inferring characteristics.
UK/EU residents: you also have the right to lodge a complaint with your local data protection authority.
10. Security
We protect information using HTTPS in transit, encryption at rest through our providers, row-level security on our database, credentials held in a managed secret store rather than in code, and access limited to the operator of the project.
No system is perfectly secure. If we become aware of a breach affecting your personal information, we will notify you and any required regulator as the law requires.
11. Children
Capitol Ledger is a public information site and is not directed at children. We do not knowingly collect personal information from anyone under 13.
Reading the site requires nothing from you. Accounts are only for people 13 or older; please do not use the feedback form or sign in if you are under 13. If you believe a child under 13 has sent us personal information or created an account, email support@capitolledger.com and we will delete it.
Ledger AI is for people aged 13 or older. If you are under the age of majority where you live, your parent or legal guardian must give you permission and agree to our Terms of Service on your behalf before you create an account or use Ledger AI.
12. Changes to this policy
We may update this policy as the project develops. The "last updated" date at the top will change, and any new category of data collection will be reflected here before that collection begins.