Privacy Policy
Plain-English summary: Index stores the memory you and your AI tools write into it. We use that data to provide the service. We do not sell your data. You can review, edit, export, and delete your saved memory.
1. What Index is and how it works
Index is a memory infrastructure service. It provides a structured store of knowledge — called Boards — that AI applications can read from and write to on your behalf. When you connect an AI tool (such as Claude, ChatGPT, or Gemini) to Index using a Connector key, that tool can push structured memories into your Boards and retrieve context when it starts a new session.
Index currently stores context you or your connected tools submit. It does not currently offer device screen monitoring.
This policy describes how we collect, use, store, and protect personal data in connection with this service.
2. Information we collect
Account information
- Email address — used for login, verification, and service communications.
- Name — optional display name used in the interface and shared Board contexts.
- Password hash — we store a scrypt hash; never the password itself.
- Account metadata — timestamps for creation, last login, email verification, and password changes.
Memory content (Boards and Entries)
- Text entries written to Boards by you or by connected AI tools on your behalf.
- Entry metadata: tags, source application, session identifiers, confidence scores, importance ratings, and access timestamps.
- Board names and descriptions you create.
Connector usage data
- Connector key metadata: name, scope, creation timestamp, and last-used timestamp.
- Write and read event logs for rate-limiting, deduplication, and security purposes.
Photos, profiles, and conversations
Content you submit can include uploaded photos, recorded chats, You and AI profile details, and context cards you save or publish. When you ask Atlas a question, relevant saved context and your question are sent to Anthropic to compose an answer.
Usage and technical data
- IP addresses — collected for rate limiting and security; stored transiently.
- Security audit events — login attempts, password changes, key creation and revocation, email changes.
- Anonymous aggregated performance metrics with no personally identifiable information.
Billing data
Payment processing is handled by Stripe. We store your Stripe customer ID and subscription status. We do not store full card numbers or payment instrument details.
3. How we use your information
- Service delivery — routing entries to Boards, deduplicating memory, serving context to connected AI tools, and generating proactive suggestions.
- Authentication and security — verifying your identity, detecting suspicious login activity, rotating tokens, and maintaining an audit log for your review.
- Communications — sending verification emails, password reset emails, and service announcements. We do not send marketing emails unless you opt in separately.
- Billing — processing subscription payments and enforcing plan limits through Stripe.
- Improvement — understanding aggregate usage patterns (not individual memory content) to improve reliability, performance, and features.
We do not use the content of your memory entries to train AI models, and we do not sell your data to third parties.
4. Memory and AI processing
Index uses large language model (LLM) APIs — currently Anthropic — to power the Orchestrator, which classifies, deduplicates, and routes inbound entries. When the Orchestrator needs LLM inference, a short excerpt of the relevant entry text is sent to the LLM API provider as part of the API request.
We use OpenAI's Embeddings API to generate vector representations of entry text for semantic search and deduplication. This means entry text is transmitted to OpenAI's API for embedding generation.
Both providers process this data under their own API data usage policies. We recommend reviewing Anthropic's Privacy Policy and OpenAI's API data usage policy.
The Proactive Engine periodically scans your Boards to identify patterns and generate suggestion cards. This processing runs within the Index backend and does not expose Board content to external AI APIs beyond what is described above.
5. Device screen monitoring
Index does not currently offer device screen monitoring. If that changes, this policy and the relevant consent flow will describe what is collected before the feature is offered.
6. Sharing and disclosure
We do not sell, rent, or trade your personal data or memory content. We share data only in these circumstances:
Sub-processors and infrastructure
- Anthropic — LLM API for the Orchestrator and memory classification.
- OpenAI — Embeddings API for semantic search and deduplication.
- Stripe — Payment processing and subscription management.
- Cloud infrastructure provider — Hosting and database services. Data is stored encrypted at rest.
Shared Boards
If you invite collaborators to a Board, members with appropriate roles can read and write entries on that Board. You control who has access to each Board.
Legal requirements
We may disclose data if required by law, court order, or government authority, or if we believe in good faith that disclosure is necessary to prevent imminent harm or illegal activity.
Business transfers
If Index is acquired, merged, or its assets are transferred, your data may be part of that transfer. We will notify users before any such transfer takes effect.
7. Data retention
- Account data — retained while your account is active until you delete it.
- Memory entries — retained until you delete them or delete your account.
- Audit events — retention varies by plan: 7 days on Free, 90 days on Pro, 365 days on Max. After the retention window, events are purged.
- Security tokens — expired and revoked tokens are purged nightly.
When you delete your account, we initiate a cascade deletion of all associated Boards, entries, connectors, and audit events. The deletion runs immediately; some operational logs (e.g. anonymised audit trails) may survive for the audit retention window above.
8. Your rights and controls
- Access — export a full copy of your data (Boards, entries, connectors, audit events) via Settings → Export.
- Correction — update your name, email address, and Board content at any time.
- Deletion — delete individual entries, entire Boards, or your entire account. Account deletion is permanent and irreversible.
- Portability — your data export is provided in JSON format.
- Objection / restriction — revoke connector keys at any time.
If you are in the EEA, United Kingdom, or Switzerland, you may have additional rights under the GDPR or equivalent legislation, including the right to lodge a complaint with your local supervisory authority. Contact us to exercise any of these rights.
9. Security
- Passwords are hashed with scrypt and never stored in plaintext.
- Access tokens are short-lived JWTs. Refresh tokens rotate on each use; presenting a revoked token invalidates the entire token family.
- Connector API keys are stored as a hash of each key, with a separate nonsecret prefix for identification; the full key is shown only once at creation.
- All data in transit is protected by TLS. Data at rest is encrypted at the storage layer.
- Rate limiting is applied to all authentication and connector endpoints.
- Security-relevant actions are logged to a user-visible audit log.
If you believe you have found a security vulnerability in Index, please contact us at [email protected] before disclosing publicly.
10. Children's privacy
Index is not directed at children under the age of 13 (or 16 in the EEA). We do not knowingly collect personal data from children. If you believe a child has provided us with personal data, please contact us and we will delete it promptly.
11. Changes to this policy
We may update this Privacy Policy from time to time. When we do, we will update the "Last updated" date at the top and, for material changes, notify you by email or by prominent notice on the dashboard.
12. Contact
Questions, data right requests, or policy concerns: [email protected]
We aim to respond to all privacy requests within 30 days.