Privacy & Security

1. Introduction

At Focus Agents, we believe in 'Managed Intelligence.' This means we provide the reasoning engine, but we do not store the fuel. This policy outlines our commitment to ephemeral, zero-retention processing.

2. Zero-Retention Architecture

Focus Agents operates on a strict 'Ephemeral Processing' model. When you deploy an agent, data (such as transcripts or Figma frames) is streamed into our secure memory environment solely for the duration of the analysis. Once the artifact (Slide Deck, PDF, or Ticket) is generated, the raw input cache is instantaneously and permanently wiped. We do not maintain a database of your intellectual property.

3. AI Model Isolation

We utilize commercial enterprise tiers of LLMs (via Anthropic & OpenAI) that explicitly opt-out of model training. Your strategic data, roadmaps, and research never become training fodder for public AI models. Your IP remains yours.

4. Permissions & Draft Mode

Our agents act as 'Drafting Partners,' not admins. For tools like Jira, our agents request only the minimum permissions necessary to stage tickets in a 'Draft' or 'Backlog' state. The agent cannot publish, delete, or alter live production data without human approval.

5. Security Infrastructure

Our managed workflows are hosted on private, isolated instances. All data in transit is encrypted via TLS 1.3. For Enterprise clients, we support deploying agents directly into your own VPC (Virtual Private Cloud), ensuring data never leaves your perimeter.