Privacy Policy

Last updated: 2026-08-05

Version: 1.0.0-mvp

This Privacy Policy describes how CoreOfDiscovery (“we,” “us”) collects, uses, and shares information when you use our website, apps, APIs, and MCP tools (the “Service”).

This is a product privacy policy for an MVP. It is not legal advice. Enterprise customers may need a DPA or additional terms.

1. What we collect

Account and organization

• Identifiers from our auth provider (e.g. Clerk): user id, email, name, organization membership and roles.
• Plan, seat, and billing metadata from Stripe (not full card numbers—Stripe processes cards).

Decision and product data

• Decisions you log: titles, descriptions, options, probabilities, barriers, tags, visibility, resolve dates, outcomes, lessons.
• Calibration and analytics derived from that data (Brier/D, DEFF snapshots, patterns).
• Reasoning trail entries, audit-log rows (including actor attribution: human vs agent).
• Workflows, kanban positions, share tokens, disagreement signals/responses.
• Optional decision-support acknowledgment (version + time) when you check the box before logging a decision.

Technical and security

• API key metadata (name, scopes, hashes)—we do not store raw key secrets after creation.
• Usage events for rate limits, x402 payment metadata (network, amounts where applicable), idempotency keys.
• Push subscription endpoints/keys if you enable Web Push.
• Server logs needed to operate and secure the Service (may include IP addresses and user agents for a limited time).

AI features

• When LLM features are enabled, prompts and responses may be processed by our model provider (e.g. OpenAI) to generate coaches, framers, path suggestions, or embeddings. We aim to minimize unnecessary retention of raw prompts where deterministic fallbacks suffice.

2. What we try not to store

• Full raw payment card data (handled by Stripe).
• Long-term raw ERP system dumps beyond what you send into decision/webhook payloads.
• Secrets in audit payloads (we design against logging API keys and passwords).

3. How we use information

• Provide decision logging, scoring, calibration, workflows, sharing, and APIs/MCP.
• Operate billing, fraud prevention, rate limiting, and abuse detection.
• Send transactional email/push you enable (resolution nudges, digests, billing notices).
• Improve reliability, safety, and product quality (including aggregated metrics).
• Comply with law and enforce Terms.

We do not sell your personal information as “data brokerage.”

4. Legal bases (where GDPR/UK GDPR applies)

Depending on activity, we rely on: contract (providing the Service), legitimate interests (security, product improvement, fraud prevention), consent (where required, e.g. certain cookies or marketing), and legal obligation.

5. Sharing

We share data with processors who help run the Service, for example:

We may disclose information if required by law, to protect rights and safety, or in a merger/acquisition with notice where appropriate.

Subprocessors for a production launch should be listed on a public subprocessors page when your domain is live.

6. Retention

• Active account: decision and account data kept while the account is open and as needed to provide the Service.
• Soft-delete: decisions may be hidden from lists but retained for a grace window (about 30 days) for restore, export, and audit continuity.
• Hard-delete: after retention, PII fields may be stripped while minimal audit breadcrumbs remain where needed for security/compliance integrity.
• Account deletion: after you request deletion and any grace period ends, we delete or anonymize personal data except where law requires longer retention (e.g. billing records, fraud logs).

Exact windows are implemented in product retention settings (`lib/retention.ts` and related jobs).

7. Your controls

Depending on region and plan:

• Export: Settings → Privacy → request an export, then download when ready (`POST /api/v1/me/export`, then `GET /api/v1/me/export?token=…`, human session).
• Delete account: Settings → Privacy (human session; agents cannot delete accounts).
• Soft-delete decisions: via product UI / API when permitted.
• LLM / Copilot: disable or limit AI features in settings where available.
• Notifications: email/push preferences and quiet hours.
• API keys: create and revoke in Settings → API keys.

To exercise rights (access, correction, deletion, restriction, portability, objection), use in-product tools first, then contact support on your production domain.

8. International transfers

We may process data in the United States and other countries where our providers operate. Where required, we use appropriate transfer mechanisms (e.g. standard contractual clauses) via those providers.

9. Security

We use industry-standard measures appropriate to an MVP SaaS product (encryption in transit, access controls, hashed API keys, audit logs). No method of transmission or storage is 100% secure.

10. Children

The Service is not directed to children under 16 (or higher age where required). Do not use the Service if you are under that age.

11. Decision-support and your responsibility

Scores and AI text can be wrong. You remain responsible for decisions you make. See the Terms of Service for disclaimers and liability limits, including the in-product acknowledgment when logging decisions.

12. Changes

We may update this Policy and will change the “Last updated” date. Material changes may be announced in-product or by email for account holders.

13. Contact

Privacy questions: support contact on the production site, or `privacy@` / `help@` on your configured domain.