Manages, stores, rotates, and injects sensitive secrets for AI agents without exposing raw values.
Sponsored
Addresses the critical security risk of exposed secrets in AI agent configurations by providing a robust Model Context Protocol (MCP) server. It encrypts secrets at rest with AES-256-GCM, issues short-lived, scoped tokens to agents instead of raw credentials, and performs server-side injection into API requests. This ensures agents never handle plaintext secrets, significantly enhancing security and providing an auditable trail of all secret access.
Características Principales
01Inject secrets directly into API request templates server-side
020 GitHub stars
03Issue short-lived, scoped tokens to agents for secure access
04Scan configuration files for exposed secrets using multiple regex patterns
05Store encrypted secrets with AES-256-GCM and rotation policies
06Maintain a full audit trail of all secret storage, token issuance, and injection events
Casos de Uso
01Securely managing API keys and other credentials for AI agents
02Preventing the accidental exposure of sensitive information in agent configurations
03Auditing and monitoring secret access to ensure compliance and detect unauthorized attempts
