Privacy Policy

This policy applies to three distinct ways people interact with AI Nazar, and the data implications differ for each:

The hosted dashboard (ainazar.com). When you create an account, sign in with GitHub, and use our hosted infrastructure to store and view your LLM usage events, we act as the data controller for the account information you give us (email, GitHub profile, organization membership) and as a data processor for the LLM event data you or your systems send to us (prompts metadata, token counts, latency, cost, error payloads, and — if you choose to enable it — request/response content).

Self-hosted deployments. AI Nazar is open source (MIT licensed) and designed to be cloned and run on your own infrastructure, backed by SQLite, Postgres, or your own Postbase project. When you self-host, we do not receive, store, or have access to any data your deployment collects. You are the data controller for that instance, and this policy does not govern how your self-hosted instance handles data — your own privacy notice to your users does. We describe self-hosting here only so you understand what the software does with data by default, so you can make informed decisions about your own deployment.

The SDK. The ai-nazar npm package wraps your OpenAI (or compatible) client and forwards telemetry about each request to whichever endpoint you configure — by default, wherever you tell it to point, including your own self-hosted instance or our hosted Service. The SDK itself does not phone home to AI Nazar unless you configure it to send data to ainazar.com.

Where this policy refers to “you,” it means the person or organization using the hosted Service, unless context makes clear we mean visitors to our marketing site.

We collect information in three broad categories when you use the hosted Service.

Account and authentication data. We use GitHub OAuth (via our authentication provider, Postbase) as the sign-in method for the hosted dashboard. When you authenticate, we receive your GitHub user ID, username, public profile name, public email address (or the email you've made available to OAuth apps), and avatar URL. We do not receive your GitHub password — authentication is handled entirely by GitHub's OAuth flow, and we never see or store your GitHub credentials.

Usage and event data. The core function of AI Nazar is to record telemetry about LLM API calls made through our SDK or sent to our ingestion API. This includes, depending on your configuration: the model name and provider (e.g., OpenAI, Anthropic, Gemini), request and response token counts, estimated cost, latency in milliseconds, HTTP status and error codes, timestamps, and an API key identifier used to attribute the event to your account. By default, the SDK does not transmit the full text of prompts or completions — it sends metadata about the call. If you explicitly enable content logging in your SDK configuration or dashboard settings, prompt and response text will also be transmitted to and stored by the Service, and you are responsible for ensuring you have the right to share that content with us, including any personal data your end users may have included in their prompts.

Technical and log data. Like most web services, our servers automatically log standard technical information when you use the dashboard or API: IP address, browser type and version, device and operating system information, referring/exit pages, timestamps, and request identifiers. We use this data for security, abuse prevention, debugging, and aggregate analytics about how the Service is used.

Information you provide directly. If you contact support, fill out a form, comment on a GitHub issue, or otherwise communicate with us, we collect whatever information you choose to include in that communication — typically your name, email address, and the content of your message.

Payment information (future paid tiers). If and when we introduce paid plans, billing will be handled by a third-party payment processor (such as Stripe). We do not store full payment card numbers on our own servers; we retain only the metadata necessary to reconcile invoices, such as plan tier, billing status, and the last four digits of a card, as provided to us by the processor.

We use the information described above for the following purposes:

To operate the Service. We use account data to authenticate you, associate events with the correct account and API key, enforce per-account rate limits and quotas, and render your dashboard views (cost charts, latency percentiles, error breakdowns, and model comparisons).

To maintain security and integrity. We use technical and log data to detect and prevent fraud, abuse, unauthorized access, and violations of our Terms of Service. This includes monitoring for anomalous API key usage, rate-limiting suspicious traffic, and investigating security incidents.

To provide support. When you contact us, we use the information you provide to respond to your question, diagnose issues with your account or integration, and follow up as needed.

To improve the Service. We analyze aggregate, de-identified usage patterns — for example, which dashboard features are used most, which providers and models are most commonly tracked, and where errors or slow page loads occur — to prioritize engineering work. We do not use the content of your prompts or completions (in accounts where content logging is enabled) to train any AI or machine learning model, ours or anyone else's.

To communicate with you. We may send you service-related emails: security alerts, changes to this policy or our Terms, confirmation of account actions, and — only if you opt in — product updates or announcements. You can unsubscribe from non-essential communications at any time.

To comply with legal obligations. We may process information as necessary to comply with applicable law, respond to lawful requests from public authorities, or protect our legal rights.

If you are located in the European Economic Area or the United Kingdom, we rely on the following legal bases under the GDPR/UK GDPR to process your personal data:

Contractual necessity. Processing your account and event data is necessary to provide the dashboard functionality you sign up for — we can't show you a cost chart without processing the events you send us.

Legitimate interests. We process technical/log data and aggregate usage analytics under our legitimate interest in securing, debugging, and improving the Service, balanced against your rights and freedoms.

Consent. Where we ask for optional content logging (full prompt/response storage) or opt-in marketing emails, we rely on your consent, which you can withdraw at any time through your dashboard settings or by unsubscribing.

Legal obligation. We process information where necessary to comply with a legal obligation we are subject to, such as responding to a valid legal process.

We do not sell your personal data. We do not share the content of your prompts or completions with third parties for advertising purposes. We share information only in the following circumstances:

Service providers (sub-processors). We use a small number of third-party vendors to operate the hosted Service, including infrastructure hosting and deployment (such as Railway), our database and authentication layer (Postbase), and, if applicable, payment processing (such as Stripe) and transactional email delivery. These providers process data on our behalf, under contractual terms that require them to protect it and use it only for the purposes we specify. We maintain a list of current sub-processors and will update it as our infrastructure evolves; you may request the current list by emailing privacy@ainazar.com.

GitHub. Because authentication is handled via GitHub OAuth, GitHub processes the authentication flow according to its own privacy policy. We only receive the profile fields GitHub's OAuth scope grants us access to.

Legal and safety disclosures. We may disclose information if required to do so by law, subpoena, or other legal process, or if we believe in good faith that disclosure is necessary to protect our rights, protect your safety or the safety of others, investigate fraud, or respond to a government request.

Business transfers. If AI Nazar is involved in a merger, acquisition, financing, reorganization, or sale of assets, your information may be transferred as part of that transaction. We will notify you (for example, via email or a notice on the Service) of any such change in ownership or control of your personal data, and of any choices you may have.

With your direction. If you invite team members to your organization within the dashboard, those team members will be able to see the event data, dashboards, and account information associated with that organization, subject to the roles and permissions you configure.

We retain account data for as long as your account is active. If you delete your account, we will delete or anonymize your account data within 30 days, except where we are required to retain it for legal, tax, or security purposes (for example, fraud investigation records may be retained longer).

Event data (LLM request telemetry) is retained according to the retention window associated with your plan. On our free hosted tier, event data is retained for 30 days on a rolling basis and then permanently deleted; paid tiers, when introduced, may offer longer retention windows, which will be disclosed at the time of purchase. Aggregated, statistical summaries derived from event data (for example, monthly cost totals) may be retained longer than the underlying raw events, as they do not identify individual requests.

Technical and log data used for security and abuse prevention is generally retained for up to 90 days, though we may retain specific records longer if they are relevant to an ongoing security investigation or legal obligation.

If you self-host AI Nazar, retention is entirely under your control via your chosen storage backend (SQLite, Postgres, or your own Postbase project) — we have no visibility into or control over that data.

We use industry-standard technical and organizational measures designed to protect the information we process, including encryption of data in transit via TLS, encryption of sensitive data at rest, access controls limiting employee access to production data on a need-to-know basis, and API key hashing so that raw API keys are not stored in plaintext in our database.

No method of transmission or storage is 100% secure, and we cannot guarantee absolute security. If we become aware of a security incident that affects your personal data, we will notify affected users and relevant authorities as required by applicable law, without undue delay.

You are responsible for safeguarding your API keys and account credentials. If you believe an API key has been compromised, you can revoke and regenerate it immediately from your dashboard settings.

AI Nazar's hosted infrastructure may process and store data in the United States and other countries where our service providers operate data centers. If you are located outside of these countries, your information may be transferred to, stored, and processed in a country that may have data protection laws different from those of your country.

Where we transfer personal data from the EEA, UK, or Switzerland to a country not deemed to provide an adequate level of protection, we rely on appropriate safeguards such as the European Commission's Standard Contractual Clauses, or equivalent mechanisms offered by our sub-processors, to ensure your data continues to be protected in accordance with this policy.

Depending on where you live, you may have some or all of the following rights regarding your personal data:

Access. You can request a copy of the personal data we hold about you. Most account and event data is already visible to you directly in the dashboard.

Correction. You can update inaccurate account information directly in your dashboard settings, or by contacting us.

Deletion. You can delete your account and associated data at any time from your account settings, or by emailing privacy@ainazar.com. Some data may be retained as described in the Data Retention section above.

Portability. You can export your event data in a structured, machine-readable format (CSV or JSON) from the dashboard.

Objection and restriction. Where we rely on legitimate interests, you may object to certain processing, and you may request that we restrict processing in certain circumstances.

Withdraw consent. Where processing is based on consent (such as content logging or marketing emails), you may withdraw that consent at any time without affecting the lawfulness of processing before the withdrawal.

Non-discrimination. We will not discriminate against you for exercising any of these rights.

To exercise any of these rights, email privacy@ainazar.com. We will respond within the time required by applicable law (typically 30 days, or one month under GDPR). We may need to verify your identity before fulfilling certain requests.

If you are in the EEA or UK and believe we have not adequately addressed your concern, you have the right to lodge a complaint with your local data protection supervisory authority.

If you are a California resident, you may have additional rights under the California Consumer Privacy Act (CCPA), including the right to know what personal information we collect, the right to request deletion, and the right to opt out of the “sale” or “sharing” of personal information. We do not sell or share personal information as those terms are defined under the CCPA.

The hosted dashboard uses a minimal set of cookies and browser storage mechanisms necessary for the Service to function:

Strictly necessary cookies. Used to maintain your authenticated session after you sign in via GitHub OAuth. Without these, you would need to sign in again on every page load.

Preference storage. We use local storage to remember interface preferences, such as your selected theme or dashboard layout, so you don't have to reconfigure them on every visit.

Analytics. We may use privacy-conscious, aggregate analytics tooling to understand how visitors use our marketing pages (for example, which pages are most viewed). Where such tools are used, we configure them to avoid collecting personally identifiable information wherever possible and to respect Do Not Track signals where technically supported.

We do not currently use third-party advertising cookies or cross-site tracking pixels on ainazar.com. If this changes, we will update this policy and provide appropriate consent mechanisms where required by law.

Because AI Nazar sits in a sensitive position — often adjacent to your application's AI traffic — we think it is worth being explicit about the things we deliberately do not do, rather than leaving you to infer them from the absence of a mention elsewhere in this policy.

We do not train models on your data. We do not use event data, logged prompts, logged completions, or any other content you submit to the Service to train, fine-tune, or evaluate any machine learning model, whether operated by us or by a third party. This is true even for aggregate, de-identified statistics — those are used for product analytics dashboards, not model training pipelines.

We do not sell personal data. We have never sold, and do not intend to sell, personal data to data brokers, advertisers, or any other third party, regardless of how “sale” is defined under applicable law in your jurisdiction.

We do not run third-party ad trackers on the dashboard. The authenticated dashboard experience (as opposed to our public marketing pages) does not load third-party advertising pixels, retargeting scripts, or cross-site tracking cookies.

We do not read your prompts for our own purposes. Even in accounts where content logging is enabled and prompt or completion text is stored, that content is displayed only to authenticated members of your own account or organization. Our engineers do not browse customer content as a matter of course; access to production data by AI Nazar personnel is limited to specific, logged circumstances such as debugging a customer-reported issue with your explicit permission, or investigating suspected abuse or a security incident.

We do not require content logging. The SDK's default configuration transmits only request metadata — never the prompt or completion text — unless you explicitly opt into content logging in your SDK initialization options or dashboard settings. You can verify this directly by inspecting the open-source SDK source code, since the wrapping logic that decides what gets transmitted is public.

We design the Service around the principle of data minimization: we try to collect only what is necessary to provide the functionality you're using. For most accounts, this means the ingestion pipeline records structured metadata (model name, token counts, cost, latency, status code, timestamp) rather than free-text content, because that metadata is sufficient to power the cost charts, latency percentiles, and error breakdowns that make up the core dashboard experience.

Where we produce aggregate reporting — for example, a running total of monthly spend across all accounts on a shared infrastructure tier, used internally for capacity planning — we compute those aggregates in a way that does not require retaining or exposing which specific account or user generated any individual data point beyond what is necessary for the computation itself, and we do not publish or share account-identifiable aggregates externally.

API keys are stored using one-way cryptographic hashing, similar to how a password would be stored, meaning that even in the event of a database compromise, raw API key values could not be directly recovered from our stored records. Only the key prefix (used for display purposes, e.g. nz_live_ab12...) is stored in plaintext so you can visually distinguish keys in your dashboard.

The Service does not make any decision that produces legal effects concerning you, or similarly significantly affects you, based solely on automated processing. Anomaly detection features (such as flagging an oversized prompt or a spend spike) are informational only — they surface a signal in your dashboard for you to review and act on; they do not automatically suspend your account, throttle your API traffic, or take any other consequential action without a human-configured rule that you yourself set up and control.

Some browsers offer a “Do Not Track” (DNT) signal or the newer Global Privacy Control (GPC) signal, which lets you indicate a preference not to be tracked across third-party websites. Because we do not use third-party cross-site advertising trackers on the authenticated dashboard, DNT largely has no additional effect there. For our public marketing pages, where privacy-conscious aggregate analytics may be used, we treat a detected GPC signal as a valid opt-out request for any processing that would otherwise be considered a “sale” or “share” of personal information under applicable state law, even though, as noted above, we do not engage in such sale or sharing today.

To give you concrete visibility rather than a vague promise of “trusted partners,” this section describes, in general terms, the categories of infrastructure providers the hosted Service currently relies on and what each category handles. We update this list as our infrastructure evolves, and material changes will be reflected in the effective date of this policy.

Application hosting and deployment. Our web application, background workers, and ingestion API run on managed cloud infrastructure. This provider has access to data only in the sense that it is stored on servers it operates; it does not use that data for its own purposes and is bound by its own data processing terms with us.

Database and authentication. Account records, API key hashes, and event data are stored in a managed Postgres database accessed through our authentication and data-access layer (Postbase). This is the same category of infrastructure available to self-hosted deployments, meaning the underlying storage technology is one you can inspect and run yourself if you want full control.

Email delivery. Transactional emails — account verification, security alerts, and (if you opt in) product announcements — are sent through a third-party transactional email provider, which processes your email address and the content of the message solely to deliver it.

Error tracking and observability (ours, not yours). Like most software teams, we use internal error-tracking tooling to catch bugs in our own application code. This tooling may incidentally capture technical context (such as a stack trace or request ID) at the moment of an error, but is configured to avoid capturing the substantive content of Your Data wherever technically feasible.

Payment processing (future). If paid plans are introduced, a PCI-DSS compliant payment processor such as Stripe will handle cardholder data directly; we will never receive or store full card numbers on our own servers.

Each sub-processor is contractually bound to use the data we share with it only to provide the specific service we've engaged it for, to implement appropriate security measures, and to not use the data for its own independent purposes, such as advertising.

If you discover a security vulnerability in the hosted Service, the SDK, or the open-source dashboard code, please report it responsibly to security@ainazar.com rather than disclosing it publicly. We ask that you avoid accessing, modifying, or deleting data belonging to other accounts while investigating a vulnerability, and that you give us a reasonable opportunity to address the issue before any public disclosure. We do not currently operate a paid bug bounty program, but we will credit researchers who report valid issues responsibly, with their permission, and we will not pursue legal action against good-faith security research conducted consistent with this policy.

In the event of a security incident that results in unauthorized access to, or acquisition of, personal data we hold, we will investigate promptly, take steps to contain and remediate the incident, and notify affected users without undue delay and, where legally required, within the timeframe mandated by applicable law (for example, 72 hours of becoming aware of a qualifying breach under the GDPR, where feasible). Notifications will describe, to the extent known at the time, the nature of the incident, the categories of data involved, the steps we have taken in response, and any steps we recommend you take to protect yourself, such as rotating an API key. We will also notify relevant regulators where required by law.

The Service may contain links to third-party websites, such as GitHub, our documentation, or blog posts referencing external tools. This Privacy Policy does not apply to those third-party sites, and we encourage you to review their privacy policies before providing them with information.

Similarly, because AI Nazar wraps your LLM provider client (such as OpenAI, Anthropic, or Google Gemini), your use of those providers remains governed by their own respective privacy policies and terms. AI Nazar does not control, and is not responsible for, how those providers process the underlying prompts or completions once they leave your application.

AI Nazar is released under the MIT license, and the full source code is available publicly on GitHub. This means anyone can inspect exactly what data the software collects and how it is processed, and anyone can run their own instance without sending any data to AI Nazar at all.

If you or your organization deploys AI Nazar on your own infrastructure — using SQLite, your own Postgres database, or your own Postbase project — all event data, account data (if you build authentication on top of it), and technical logs stay entirely within your infrastructure. We have no access to, visibility into, or copy of that data. You are solely responsible for complying with applicable privacy laws with respect to your own deployment, including providing your own privacy notice to your end users if your deployment processes their personal data.

We may collect limited, aggregate telemetry about software versions and anonymous installation counts for self-hosted instances only if such a feature is explicitly documented in the project's README and can be disabled via configuration. As of the effective date of this policy, no such telemetry is implemented in the self-hosted distribution.

United States state privacy laws. In addition to the CCPA rights described above, if you are a resident of a state with its own comprehensive privacy law — including Virginia, Colorado, Connecticut, Utah, Oregon, Texas, or others that have since enacted similar frameworks — you may have rights to access, correct, delete, and port your personal data, and to opt out of targeted advertising, the sale of personal data, and certain profiling. As described throughout this policy, we do not engage in the sale of personal data or targeted third-party advertising on the authenticated dashboard, and we do not conduct profiling that produces legal or similarly significant effects. To exercise any state-specific right, email privacy@ainazar.com; we will honor authorized-agent requests where permitted by the relevant state law, subject to identity verification.

Canada. If you are located in Canada, we process personal information consistent with the Personal Information Protection and Electronic Documents Act (PIPEDA) and applicable provincial equivalents. You may direct inquiries or complaints about our handling of your personal information to privacy@ainazar.com in the first instance.

Brazil. If you are located in Brazil, we process personal data consistent with the Lei Geral de Proteção de Dados (LGPD), relying on legal bases analogous to those described in the “Legal bases for processing” section above, and you may exercise LGPD rights (confirmation, access, correction, anonymization, portability, deletion, and information about sharing) by contacting privacy@ainazar.com.

Australia. If you are located in Australia, we handle personal information in a manner consistent with the Australian Privacy Principles under the Privacy Act 1988 (Cth), including taking reasonable steps to keep personal information accurate, secure, and used only for the purposes disclosed in this policy.

Where any region-specific law grants rights broader than those described elsewhere in this policy, we will honor the broader right for residents of that region, even where this section does not enumerate it exhaustively. If you are unsure which rights apply to you, email privacy@ainazar.com and we will help clarify.

This Privacy Policy should be read together with our Terms of Service, which governs your broader use of the Service, including account eligibility, acceptable use, intellectual property, and liability. Where the Terms of Service discuss “Your Data,” that term has the same meaning as the event data, content, and account information described in this Privacy Policy. In the event of a direct conflict between a data-handling commitment in this Privacy Policy and a provision of the Terms of Service, this Privacy Policy controls with respect to the processing of personal data, and the Terms of Service control with respect to all other matters, such as liability and dispute resolution.

The Service is not directed to individuals under the age of 16, and we do not knowingly collect personal data from children. AI Nazar is a developer tool intended for use by software engineers and organizations building AI-powered applications. If we become aware that we have collected personal data from a child without verified parental consent, we will take steps to delete that information promptly. If you believe a child has provided us with personal data, please contact us at privacy@ainazar.com.

We may update this Privacy Policy from time to time to reflect changes in our practices, technology, legal requirements, or other factors. When we make material changes, we will update the “Effective date” at the top of this page and, where required by law or where the changes are significant, provide additional notice such as an email to registered account holders or a prominent notice on the dashboard.

We encourage you to review this policy periodically. Your continued use of the Service after a revised policy becomes effective constitutes your acknowledgment of the updated terms, to the extent permitted by applicable law.

If you have questions, concerns, or requests regarding this Privacy Policy or our data practices, please contact us at:

Email: privacy@ainazar.com
GitHub: github.com/harshalone/ai-nazar

We aim to respond to all privacy-related inquiries within 5 business days, and to fulfill verified data subject requests within the timeframes required by applicable law.