Legal Framework

Privacy Policy

Last updated: June 24, 2026

1. Introduction

ContextMiner ("we", "our", or "us") respects your privacy and is committed to protecting your personal data. This privacy policy explains how we collect, use, and safeguard your information when you use our AI gateway, token optimization, and proxy services.

2. Information We Collect

2.1 Account Information

  • ↳Email address (used for secure authentication, billing, and transactional communication).
  • ↳Name or organization details (optional, for personalization).
  • ↳Passphrase credentials (securely stored in hashed format using robust key-derivation algorithms).
  • ↳Billing coordinates and subscription status (processed securely through Stripe integration).

2.2 Upstream Model API Keys (BYOK Model)

To route prompts to downstream AI models, we collect your provider API credentials. We enforce strict data protection controls:

  • ↳API keys are encrypted at rest using AES-256 with unique Initialization Vectors (IV).
  • ↳Keys are decrypted dynamically and strictly in-memory during the proxy request lifecycle. Plain text keys are never committed to permanent storage, database tables, or execution logs.
  • ↳Keys are deleted immediately from all database tables upon user request or account deletion.

2.3 Aggregate Usage Data

  • ↳Request volumes and timestamps (excluding payload body).
  • ↳Token consumption metrics (original vs. clean token counts).
  • ↳Downstream provider routing identifiers and model tags.
  • ↳Estimated cost savings calculations.

3. How We Use Your Information

We process the data we collect for the following operational purposes:

  • ↳Providing, operating, and maintaining our API proxy gateway.
  • ↳Executing real-time token reduction and PII masking.
  • ↳Generating analytics dashboards showing cost savings and volume metrics.
  • ↳Managing your account status, billing cycles, and subscription tiers.
  • ↳Communicating developer alerts, maintenance windows, and policy updates.
  • ↳Improving our Structural Logic Analyzers and Web Noise Filtering selectors.

4. Data Ingestion & Proxy Processing

4.1 Transient In-Memory Processing

ContextMiner acts as an in-memory routing proxy. When you send prompt messages through our completions endpoint:

  • ↳All compression engines (including our Dynamic Context Optimizer and Structural Logic Analyzers) and NLP-based Entity anonymizers operate strictly in-memory within our self-contained, private infrastructure. No prompt payloads, code, or context are ever transmitted to third-party telemetry, external classifiers, or AI training sets.
  • ↳Prompt text content is processed strictly in volatile memory to execute Structural Logic Analysis, Web Noise Filtering, log deduplication, and zero-egress entity anonymization via our Token Vault.
  • ↳Anonymized and optimized prompts are immediately forwarded to downstream model providers.
  • ↳Under our Media Block Policy (General Mode), multimodal payloads (images, audio, video) are strictly blocked at ingestion to prevent visual PII leaks and guarantee low proxy latency. In IDE/Dev Mode, multimodal payloads are fully allowed.

4.2 Data Retention Policy

ContextMiner is designed with zero-egress, trust-free principles regarding prompt data retention:

  • ↳We do NOT store or log the text content of your prompts or AI provider responses. Once routed, prompt payloads are discarded from memory.
  • ↳We only persist aggregated numeric metrics (byte length, token count details, savings calculations) for billing and analytics dashboards.
  • ↳We retain account settings and encrypted API keys only while your profile remains active.

5. Data Sharing and Third-Party Transfer

We do NOT sell, lease, or monetize your prompt data or personal information. We only share information with:

  • ↳AI Providers: Optimized prompts are routed to the specific AI provider endpoint (OpenAI, Anthropic, Gemini, DeepSeek, and others) using your BYOK credentials.
  • ↳Stripe: Subscription billing details are processed securely by Stripe under their privacy policies.
  • ↳Compliance and Regulatory Authorities: Only if strictly required under a subpoena or court order.

6. Data Security and Custody Controls

We implement industry-standard security structures to safeguard your API keys and profile information:

  • ↳AES-256-GCM encryption for API keys at rest.
  • ↳SSL/TLS encryption for all incoming and outgoing network data transfers.
  • ↳Database access controls limited strictly to internal proxy components.
  • ↳Regular security auditing and vulnerability scanning.

7. Your Rights and Data Portability

Under global regulations (GDPR, LGPD, CCPA), you hold the following rights:

  • ↳The right to access and export your account settings and usage histories.
  • ↳The right to correct or update account details and model key credentials.
  • ↳The right to request immediate deletion of your profile, encrypted keys, and billing references.
  • ↳The right to revoke consent for proxy data processing (which terminates service access).

8. Cookies and Tracking Technologies

We use minimal cookies for authentication sessions and preference caching. We do not deploy cross-site marketing cookies or tracking pixels.

9. Third-Party Integrations

ContextMiner integrates with the following processors:

  • ↳Stripe: Handles recurring billing. Review their privacy policy at https://stripe.com/privacy.
  • ↳PostgreSQL: Stores encrypted keys and usage metrics securely under local hosting configurations.
  • ↳AI Providers: OpenAI, Anthropic, Google, DeepSeek (as configured by your requests).

10. Children's Privacy

Our service is not intended for users under 18. We do not knowingly collect data from children. If you believe a child has created an account, please contact us immediately for deletion.

11. International Data Transfers

Our servers operate in the United States. If you access our services from the European Union, Brazil, or the UK, your data (limited to account and billing references) will be processed in accordance with local regulations, including GDPR and LGPD standards.

12. Policy Changes

We reserve the right to update this privacy policy at any time. Changes will be posted with an updated date below. Significant changes will be communicated via email or dashboard notifications.

13. Contact Information

If you have any questions, concerns, or privacy inquiries regarding this Privacy Policy or our local data anonymization engine, please contact us directly: