Frequently Asked Questions

Absolutely. Data security is foundational to how we built Cora:

• Your infrastructure, your data: Agents are securely hosted on AWS infrastructure through our partnership. Enterprise-grade security and compliance built in.
• Encryption everywhere: All data is encrypted at rest and in transit using industry-standard AES-256 and TLS 1.3.
• No training on your data: We never use your business data to train AI models. Your data is yours alone.
• Daily security audits: Automated security audits check for anomalous access patterns and configuration drift.

We follow a three-tier progressive autonomy model:

• Tier 1 (Reads): Data retrieval, analysis, searches. Runs freely with no approval needed.
• Tier 2 (Updates): Reversible modifications like updating spreadsheets or drafting documents. Soft confirmation required.
• Tier 3 (Sends/Deletes): Irreversible external actions like sending emails or posting messages. Hard approval required.

All agents start at Tier 3 for everything. Actions graduate to lower tiers only after 100+ correct outputs with zero corrections. You control when autonomy expands.

Yes. Complete audit logging is built into every deployment:

• Every action is logged with timestamp, context, and outcome.
• You can see what the agent did, when it did it, and why.
• Logs are exportable for compliance and review.
• Real-time activity monitoring via dashboard or Slack/Discord.

We take compliance seriously and designed Cora to meet enterprise requirements:

• Security controls: Infrastructure and processes built for enterprise audit readiness.
• Healthcare compliance: Tailored data-handling configurations and documentation support for healthcare customers.
• GDPR: Full GDPR compliance including data portability and right to deletion.
• AWS hosting: Securely hosted on AWS infrastructure through our ISV partnership with enterprise-grade isolation.

Need specific compliance documentation? Contact us.

AI isn't perfect, and we've built Cora with that reality in mind:

• Configurable guardrails: Set limits on what actions require approval (e.g., emails, external communications, financial actions).
• Undo and rollback: Most actions can be reversed, and every action is logged for easy correction.
• Escalation paths: Configure when the agent should stop and ask a human.
• Learning from corrections: When you correct the agent, it remembers and improves.

The goal isn't perfection. It's to be dramatically better than the alternative: work not getting done, or hiring and training for months.

When an agent encounters an error or unexpected situation:

1. Automatic retry: For transient errors (API timeouts, network blips), it retries intelligently.
2. Alternative approaches: If one method fails, it tries other ways to accomplish the goal.
3. Escalation: If it can't resolve the issue, it notifies you via your preferred channel (Slack, Discord, email, SMS).
4. Documentation: Every error is logged with full context for review.

You can configure how aggressive vs. cautious you want error handling to be.

ChatGPT is a chatbot you talk to. Cora enables the agentic-first business. The difference is fundamental:

ChatGPT: Answers questions when asked. Requires constant prompting. Can't take actions in the real world. Forgets context between sessions.

Cora: Builds a persistent brain containing your institutional knowledge. Deploys agents that take real actions across your systems. Remembers everything. Runs 24/7. Compounds intelligence over time.

Think of ChatGPT as a search engine you talk to. Cora is the operating system for your business intelligence.

The agentic on-ramp has an entry point for every stage of growth:

• Team Brains (coming soon): Shared skills and blessed policies synced from Google Drive, Notion, and Process Street. Permission-controlled access so everyone sees only what they should. Model-agnostic. 3 free brains included.
• Brain Boot: 2-hour session to build your brain with Claude Code and Obsidian. Hundreds of skills installed.
• Agents: Powered by Claude Code, Codex, or Gemini. An agentic computer that can do any job in your business. 24/7 support, community access, and continuous upgrades.
• Executive Brain Coaching: Six months of direct coaching with Vinay plus Forward AI engineering support.
• AI Transformation: Full enterprise rebuild over 12 months. Department-by-department migration.

Services create the brain. Software keeps it running. See pricing for full details.

You have complete control. Agents are designed with governance as a core principle:

• Approval workflows: Require human approval for specific action types.
• Access controls: Define exactly which systems and data the agent can access.
• Time boundaries: Set when it can and can't work or contact people.
• Spending limits: Cap financial actions or require approval above thresholds.
• Kill switch: Pause or stop an agent instantly from any device.

Start conservative and expand permissions as trust builds.

Cora isn't about replacing your team. It's about augmenting their capacity:

• Your existing team focuses on high-value, creative, strategic work.
• Tasks that weren't getting done now get done.
• Coverage gaps (nights, weekends, holidays) are filled.
• New hires can be delayed or avoided entirely.

Most customers use Cora to handle work that was either falling through the cracks or required hiring someone they couldn't justify.

Agents connect to thousands of apps and orchestrate your entire stack. No more drag-and-drop workflows. No more manual integrations. The agent, the brain, and your tools combine into one system.

• MCP connections: Thousands of apps connected. CRMs, support tools, project management, accounting, and more.
• APIs: Direct API integrations for faster, more reliable automation.
• Web and desktop apps: Agents can navigate UIs like a human would.
• Email: Gmail, Outlook, or any IMAP/SMTP provider.
• Communication: Slack, Discord, Teams. Agents can be messaged directly.

The agent orchestrates everything with your brain as context. You never build another integration again.

A brain is a structured collection of markdown files, Python scripts, and other scripts organized in the ACE folder structure (Atlas, Calendar, Efforts, Inbox, x_System):

• Each brain syncs with a GitHub repo. The repo is the universal interface for any agent to access the brain.
• External sources sync one-way into the brain: Process Street Pages, Google Docs, Notion.
• Brains can be for people (personal second brains), agents (purpose-built), or roles (templates).
• The brain contains skills (reusable AI workflows), templates, and institutional knowledge.

The brain is the intelligence layer that makes agents effective for your business.

All agents are securely hosted on AWS infrastructure through our ISV partnership:

• AWS hosting: Enterprise-grade cloud infrastructure with AI inference powered by AWS Bedrock.
• Managed operations: We handle hosting, uptime, security, and monitoring.
• AWS Marketplace: Available for streamlined enterprise procurement.

We are an AWS GenAI Competency certified partner. Your agents run on secure, managed infrastructure with full audit trails and compliance built in.

The brain learns through a combination of methods:

1. Brain boot: The initial 2-hour session builds your brain structure, installs skills, and connects tools.
2. Documentation sync: Google Docs, Notion, Process Street Pages sync into the brain automatically.
3. Custom skills: We build skills specific to your workflows (each is tested against real data).
4. Feedback loops: When you correct the agent, it adapts. Corrections compound over time.
5. Continuous improvement: The brain gets smarter every day as it accumulates knowledge and refines skills.

All products feed into a unified platform:

• Brain Boot is the entry point. Gets a working brain and one system audit.
• Executive Brain Coaching is for leaders who want ongoing guidance. Brain boot included.
• AI Transformation is the full enterprise engagement. Includes everything.
• Team Brains (coming soon) is the recurring infrastructure all brain users need.
• Agents is the recurring execution layer for agent-first operations.

Services create the brain. Software keeps it running. Additional consulting available.

You can cancel anytime with no penalties:

• SaaS products (Team Brains, Agents) are month-to-month.
• All your data is exported before termination.
• You own your brain, your data, and your infrastructure. No vendor lock-in.
• Open-source technology means you can continue operating independently.

We want you to stay because Cora delivers value, not because you're trapped.

Included: Everything described in each product tier. No hidden fees.

Not included (all products):

• AI model token costs (billed to your accounts).
• Hardware costs if applicable.
• Third-party software subscriptions (Apollo, Hunter, etc.).

Additional consulting and skill building is available on request.

Most customers compare an Agent to a full-time hire:

• No recruiting, interviewing, or onboarding delays.
• Works 24/7 without overtime or burnout.
• Expert-level execution across multiple domains.
• No benefits, taxes, or HR overhead.
• Can switch roles overnight.

A junior SDR costs $60,000-$80,000/year fully loaded and takes 3-6 months to ramp. A Cora agent is productive on day one.