AI AGENT SECURITY
ストックにはログインが必要です
Block dangerous AI agent actions before execution
Artificial Intelligence
Developer Tools
GitHub
Open Source
Security
Most AI security tools filter prompts before they reach the model. Relay checks the action instead. It sits between your agent and its tools, validating GitHub calls, shell commands, database queries, API requests, MCP tools, and deploys before execution. Risky actions are blocked, safe actions continue, and every decision is logged.
投票数: 3