PQVault FAQs

Question 1

How do AI agents access secrets in PQVault?

Accepted Answer

AI agents interact with PQVault via the Model Context Protocol (MCP) using specialized tools like `vault_proxy`. Instead of retrieving the key directly, agents request PQVault to proxy the API call on their behalf, with the key automatically and securely injected, ensuring the secret never leaves the vault process.

Question 2

What makes PQVault different from other secrets managers?

Accepted Answer

Unlike traditional or enterprise secrets managers, PQVault is tailored for AI agent workflows. It offers zero-knowledge API proxy, seamless MCP protocol integration, per-key rate limiting, usage tracking, and post-quantum encryption, all in a single, free, and open-source binary requiring no complex infrastructure.

Question 3

What is PQVault?

Accepted Answer

PQVault is a post-quantum secure secrets management tool specifically designed for AI agents. It provides encrypted storage, zero-knowledge API access, and granular control over API keys, ensuring your agents' access is both secure and auditable.

Question 4

Can I track and control AI agent API usage with PQVault?

Accepted Answer

Absolutely. PQVault provides comprehensive features for per-key rate limiting (RPM, daily, monthly), detailed usage tracking, and cost estimation. This allows you to monitor which agents are consuming resources, prevent budget overruns, and maintain a full audit log of every API access.

Question 5

How does PQVault secure AI agents' API keys?

Accepted Answer

PQVault employs hybrid post-quantum encryption (ML-KEM-768 + X25519 + AES-256-GCM) and a unique zero-knowledge API proxy. This architecture ensures that sensitive API keys are never directly exposed to the AI agent, even during use, maintaining a high level of security against current and future threats.

PQVault

PQVault

Key Features

Use Cases

Key Features

Use Cases