Security Overview

AeroFTP is designed with security as a foundational requirement. Every layer of the application follows a defense-in-depth model where multiple independent controls protect against any single point of failure.

Architecture at a Glance

Layer	Purpose	Details
Credential storage	Protect secrets at rest	AES-256-GCM vault, OS keyring, Argon2id KDF, optional TOTP 2FA
Encryption	Protect user files and data in transit	AeroVault v2 (AES-256-GCM-SIV), rclone crypt interoperability, TLS/SSH, archive encryption
AI tool security	Control AI agent capabilities	Backend grant system, native OS dialogs, tool classification
Supply chain	Verify software authenticity	Sigstore signing, client-side verification, update helper hardening
Privacy	Minimize data footprint	Zero telemetry, local-only storage, memory zeroization

Each layer operates independently. A compromise in one does not cascade to the others.

Trust Boundaries

The Rust backend is the security authority. The web frontend can request operations but cannot bypass backend enforcement:

• Credentials: The frontend never handles raw passwords or tokens. The backend resolves them from the vault internally.
• AI tools: Mutative operations require a cryptographic grant issued and validated by the backend, confirmed via native OS dialogs that the web layer cannot suppress.
• Updates: The backend verifies Sigstore bundles before allowing any installation. On Linux, the privileged helper re-verifies SHA-256 before executing package managers.
• Compatibility crypto: rclone crypt decryption happens locally in the backend so encrypted remotes can be inspected without handing passwords to external services.

Where AeroFTP Stands

In the landscape of file transfer and cloud management tools, AeroFTP combines protocol breadth (7 transport protocols plus native provider integrations) with security depth typically found only in dedicated encryption tools:

• OpenSSF Best Practices: 100% passing criteria
• Aikido Security: Top 5% benchmark, 0 open issues (March 2026)
• 300+ security findings identified and resolved across 9 independent audits
• Zero unresolved CVEs in application code

Reporting Vulnerabilities

Report security issues via GitHub Security Advisories. We respond within 48 hours. See the full disclosure policy and bug bounty program.