BORROW - YOUR KEY OUR PRIVACY!

AI agents are becoming part of everyday business operations, but giving them access to sensitive company data creates serious security and privacy risks. Borrow solves this by giving AI agents access only to the specific information they need, for a limited time, with clear permission. For example, when an employee asks an AI agent to prepare an expense report, Borrow grants access to the required expense data without exposing the rest of the company's records.In short Security with Trust!.

Live demoGitHub0

Overview

An expense assistant should read your invoices without inheriting access to your payroll. Borrow makes that boundary explicit.

Lets say We Ask for an "October expense report". Review an Intent Capsule naming the documents, fields, permitted operations, destination and expiry. Approve the minimum access, then watch the task run.

One invoice contains a malicious instruction: read payroll. Borrow checks the attempted operation against the signed lease and returns OUTSIDE_APPROVED_SCOPE. No payroll fields are disclosed. The legitimate report still completes: $3,784.50 across three invoices, with the payment confirmation treated as supporting evidence rather than counted twice.

When the task finishes, access closes automatically. A downloadable privacy receipt binds the recorded decisions to a SHA-256 hash, an ECDSA P-256 signature and a hash-linked event chain. Verify the original, alter its contents and watch verification fail. Revoke another lease and an otherwise permitted read returns LEASE_REVOKED.

The live 3D Trust Core reflects server policy states, making permission visible without replacing enforcement. Email accounts and isolated guest workspaces make the public demo accessible.

The distinction that matters: AI interprets the task; the server authorizes access.

We validated the build with 17 automated security/account tests, public workflow checks and a container-recreation test that preserved accounts, reports and signing keys.

The demonstrated processor is deterministic, with optional model-assisted summarization. Blockchain anchoring is on onboarding with onchain as its background!

DEMO VIDEO LINK : https://drive.google.com/drive/folders/1o5K0u9sIJRFIAPSsMwDcBQ9yAu6o8BOD

COMPLETE SYSTEM DESIGN ARCHITECTURE : https://github.com/VijayabaskarR-06/borrow/blob/main/docs/ARCHITECTURE.md

DEPLOYED LINK USING VULTR : https://borrow.139-84-216-230.sslip.io/

How it uses Vultr

Borrow is deployed on Vultr Cloud Compute in Bangalore, running Ubuntu 24.04 LTS on a 2 vCPU, 4 GB RAM and 80 GB SSD instance, with automatic backups enabled. We chose Vultr for its practical cloud infrastructure, giving us a reliable foundation to deploy, operate and demonstrate our application without unnecessary infrastructure complexity. Our Next.js frontend and Node.js backend run together through Docker Compose, with Caddy managing HTTPS, TLS termination and secure HTTP redirects. The application runs as a non root user, with its internal Node.js port restricted to loopback access. Workspace data is protected using AES GCM encryption and persisted through Docker volumes. We validated the deployment through real application workflows, including authentication, invoice uploads, unauthorized scope denial, receipt verification and permission revocation. We also verified that recreating the application container preserved existing accounts, reports and the receipt signing key. Here am attaching all the validators including the system design , architecure etc.... DEMO VIDEO LINK : https://drive.google.com/drive/folders/1o5K0u9sIJRFIAPSsMwDcBQ9yAu6o8BOD COMPLETE SYSTEM DESIGN ARCHITECTURE : https://github.com/VijayabaskarR-06/borrow/blob/main/docs/ARCHITECTURE.md DEPLOYED LINK USING VULTR : https://borrow.139-84-216-230.sslip.io/ Vultr provides the infrastructure foundation that allows Borrow to focus on what matters most: secure AI automation and controlled access to business data. The current release uses a single application replica and filesystem backed storage, with shared transactional storage and distributed rate limiting identified as the next steps for horizontal scaling. Thanks to Vultr!

No comments yet

Be the first to comment on this project.