Skip to content
Hoot Lens.Start free

Privacy

How Hoot Lens, provided by Parallel Platforms, LLC, handles account information, website recording evidence, connected agents and visits to our own public website.

Last updated

Markdown copy

Accounts and sign-in

We use Firebase Authentication to sign you in and maintain your account identity, including your account ID, email address, display name when supplied, email verification and sign-in metadata. We use this information to authenticate requests, enforce site access and help you manage your account.

This browser can remember accounts for switching between sign-ins. Its saved account list contains account IDs, email addresses, names when available and last-used times. Firebase keeps the sign-in state separately in browser storage. Remove saved accounts or clear this site’s browser data when you no longer want them remembered, especially on a shared device.

When a new account is created, we can record its signup flow (website, connected application, agent setup, or team invitation), original sign-in method, referring website hostname, and first recognized public page on Hoot Lens. Referrer and page metadata omit query strings, fragments, and private account paths. Browser privacy signals and declined recording choices suppress that referral metadata. A connecting application can supply its name; that name is not independently verified. These records help authorized administrators understand acquisition and support setup. Older accounts may have no signup-origin record.

If you answer the optional ‘How did you find Hoot Lens?’ question, we store that answer with your account to understand acquisition. You can edit or clear it. Authorized business administrators can review account identity, owned sites, current plan, setup milestones and measured service activity to operate and support Hoot Lens.

Customer sites and recording data

Customers configure which pages Hoot Lens records and the applicable privacy, consent, masking, sampling and retention settings. The service processes site configuration, captured page structure and permitted content, interaction events, recording evidence, derived counts and customer notes to provide replay, heatmaps, findings and other analysis.

Recording content varies with those settings and may include personal information visible on the recorded page. Privacy controls reduce capture; they do not make every recording anonymous or replace the site owner’s notices, permissions and legal responsibilities. Owners must keep private and sensitive content outside capture.

When a customer enables replay asset preservation, we copy supported public images, stylesheets, and fonts referenced by permitted recordings into private storage. Copies follow the recording’s retention and deletion controls. The worker fetches eligible files from registered site origins and customer-approved public asset origins without visitor cookies, authorization, or Referer headers. Existing sites keep live-asset behavior until preservation is enabled. Coverage and capture timestamps identify incomplete or late copies.

Customers can configure business outcomes reported by their browser or backend after an operation succeeds. These events use a static name, an opaque receipt identifier, anonymous recording and visit identifiers, a timestamp, and optionally a monetary amount and currency. Arbitrary customer properties and form contents are not accepted in this event interface. Customers must keep personal information out of event names and identifiers. Reports describe the outcomes the customer’s integration sent and distinguish browser from server reporting.

We store service data using Google Cloud and Firebase. A customer’s authorized members and agents can access the evidence their permissions allow. Shared recording or heatmap links can expose the selected evidence to people with access to those links. Share them only with intended recipients. The recorder documentation describes capture protections and the settings that change what is recorded.

Connected agents and APIs

Connecting an assistant through OAuth creates an authorization record for that client. We store the client identity, your account association, approved scopes, selected site IDs or account-wide site access, the target service and creation, use and revocation information to enforce and manage the connection. Connection secrets and tokens belong in secure credential storage, not recorded page content.

An authorized assistant can receive your account ID and email address, permitted site identities, roles and scopes, and the selected site evidence returned by the tools it calls. Read access can include captured element labels and evidence. Raw replay requires the replay scope. Narrative tools can return recording and visit identifiers, recorded text labels, interaction evidence and customer notes with the read scope when the authorizing account has permission to watch replays on that site. Grant only the access needed and review what the assistant requests.

That information leaves Hoot Lens for the connected assistant or service, including OpenAI when you connect ChatGPT. Its handling is subject to that provider’s terms, privacy practices and retention choices. This is separate from Hoot Lens’s own optional generated explanations described below.

Revoke an OAuth connection in Connected apps, or revoke a personal access token through its dashboard control, to stop future authorized access. Revocation does not retrieve information already sent to the provider or delete your Hoot Lens account and site data. The API and agent reference explains scopes and connection controls.

Optional generated explanations

When a site has generated explanations enabled and a configured model is used, Hoot Lens sends a bounded summary of permitted aggregate snag data to OpenAI. It can include counts, device class, sanitized page paths, release and variant labels, selectors and filtered element labels when the site’s privacy setting allows them.

This backend explanation request does not send raw recordings, DOM page text, replay events, visitor identifiers, visit or recording IDs, or the owner’s identity. It is a different data flow from giving an external assistant replay access. The owner can turn explanations off in site settings. The explanation panel shows its summary under What was sent.

Explanations describe observed evidence and may be incomplete or wrong. They do not establish a visitor’s thoughts or feelings. OpenAI’s handling of these requests follows the applicable API terms; this page does not promise a particular provider retention period.

Billing, email and support

When billing is enabled and you subscribe, Stripe processes payment information. Hoot Lens stores billing identifiers, subscription and plan status, trial and billing-period information so we can apply your plan and provide billing controls. We do not store full payment-card details in Hoot Lens.

When email features are enabled, Resend receives recipient addresses and message content to send account verification, password reset, welcome, billing or site notifications. Service records track processing outcomes to manage retries. Eligible replies to a welcome email may become feedback issues associated with the recipient’s account. Support mail can be forwarded to the Hoot Lens team.

Support and feedback records include the information you submit and the team’s replies. Do not send passwords, access tokens or private visitor content in support messages. Authorized staff can use these records to investigate and respond to issues.

Service measurement

We record daily account activity markers for successful authenticated site-data reads, including whether a request used a personal access token or OAuth. These markers include the account ID, day and observation time and help us understand account activity and agent API adoption.

We also record account-bound connection steps, such as opening or declining an approval, approving an app, completing app sign-in, an attributable sign-in failure, and observed connection use. These records contain step times and limited app identity information to help authorized administrators diagnose setup problems. They do not contain tokens, authorization codes, one-time confirmation values, or visitor recording content. Earlier connection steps may not have been measured.

These measures do not prove that a unique person used a screen or that an MCP tool completed successfully. Package downloads are a separate signal. Acquisition reports use optional signup answers, current site setup and current subscription status; they do not establish historical revenue or why someone converted.

Operational and security logs

We use operational and security logs to diagnose failures, protect the service and investigate access. Infrastructure request logs can include connection IP addresses and request metadata. These logs are separate from visitor recording data; not storing an IP address in a recording does not mean the cloud infrastructure keeps no address information.

Our configured Google Cloud Logging retention is 30 days for ordinary service logs and 400 days for required cloud audit logs. These are log-retention settings, not a promise that every service record is deleted on those dates. Google, Firebase, Stripe, Resend and connected assistant providers also apply their own terms and privacy practices to the information they process.

Connection and activity retention

Credential expiry stops its authorized use separately from removing stored records. Active OAuth grants have no fixed automatic purge time while they remain active. You can revoke them or contact support about deletion.

The following periods make records eligible for automatic cleanup. Cleanup runs asynchronously, so physical deletion can happen later than eligibility and is not an exact deadline.

  • OAuth authorization requests: one hour after the request expires.
  • OAuth authorization codes and token records: one day after expiry.
  • Inactive OAuth client records: 90 days after their last use.
  • Revoked OAuth grants: 30 days after revocation.
  • Connection-step records: 90 days after the latest recorded step. Active grant records can retain approval, sign-in completion, and use information separately.
  • Daily account-activity markers: 90 days after the observation.

Retention and your choices

Customer recordings follow their effective configured retention and plan limits. Changing retention or plans can shorten how long they remain available. Use available export and deletion controls while the data is accessible. Deletion and expiry do not guarantee that a copy already shared with someone else can be recovered or removed.

Deleted recording objects can remain as soft-deleted recovery copies in Google Cloud Storage for up to seven days after deletion. They are unavailable through Hoot Lens during that window. Deleting a recording from the service is therefore separate from immediate physical erasure of every storage copy.

Saved account entries persist in browser storage until removed, displaced by the saved-account limit or cleared with the site’s browser data. Account profiles, billing profiles, support issues and signup-email processing records have no fixed automatic deletion period. They may remain to operate the service, resolve requests, prevent duplicate processing or meet legal obligations. Expiring a credential, revoking access or canceling a subscription is separate from deleting these records.

To ask about access, correction, export or deletion of your Hoot Lens account and service data, contact support@hootlens.com. We may need to verify your identity and authority over the data. We do not promise immediate deletion of every record. Applicable privacy rights remain available.

If you are a visitor to a customer’s website, contact that site’s owner about its recording practices and your data. For visits to hootlens.com itself, the recording controls and disclosure below apply.

Visit recording on this website

We use Hoot Lens, our own product, to see how visitors use the public pages of hootlens.com: the home page, the developer FAQ, the example workspace and this page. We use what we learn to improve those pages and the sign-up flow. Signed-in workspaces, shared replay and heatmap links, invitation pages and install links are never recorded.

  • What is recorded: Page text, images, and styling are recorded as visitors saw them. Form inputs are masked.
  • Where visits come from: the name of the referring site, campaign tags and the name of an ad network from the landing address (never the ad click ID), the name of an in-app browser such as Facebook, and the country and region (state or province) of the visitor. Location is looked up from the connection and the address is not stored.
  • Consent: Visitors are recorded on load, except in European Economic Area, United Kingdom, Switzerland, Washington State, Nevada, Connecticut, where recording starts only after the visitor agrees. Global Privacy Control and Do Not Track stop recording.
  • Never recorded: password fields, payment card fields, one-time-code fields, anything the site marks private, text after # in page addresses, the values typed into form fields.
  • Page addresses: query strings are removed.
  • Links that leave a page: the page it led to on this site, or only the name of another site. Never a query string or text after #.
  • Share of visits: 100%.
  • Retention: recordings are deleted after 30 days.

In the United States, recording is on by default for most visitors. Visitors in the European Economic Area, the United Kingdom, Switzerland, Washington, Nevada and Connecticut, and visitors whose location we cannot tell, are asked first, and nothing is recorded until they agree.

You can turn recording off or on at any time with the Visit recording control in the footer of each page. Global Privacy Control and Do Not Track are honored automatically.

To show where visits come from, we record the name of the site that referred a visit, campaign tags in the landing address, the name of an ad network when an ad click ID is in the address (never the ID itself), the name of an in-app browser such as Facebook, and the visitor's country and region. Country and region are looked up from the connection when the visit starts. The lookup IP address, city and coordinates are not stored in recording data. Visitors who send Global Privacy Control or Do Not Track are not recorded or located.

The address handling described here applies to recording data. Infrastructure request and security logs are separate and can contain connection IP addresses, as described under Operational and security logs above.

IP geolocation by DB-IP.

Recordings are stored with Google Cloud. Aggregated counts and statistics, never recordings, may be sent to OpenAI to write explanations of what the data shows.

The aggregate-only explanation processing described here is separate from an account holder authorizing an external assistant to read site evidence. Connected agents can receive the recording content their approved permissions allow, as explained above. Directory badges on our homepage load images from the listed directory, which receives the connection IP address and browser request headers. These image requests are separate from Hoot Lens recordings.