Launching today
BLAZLE
In-flight security for AI agents
1 follower
In-flight security for AI agents
1 follower
Stop prompt injection, secret leaks and risky MCP tool calls before they leave your AI agents. Works with Cursor, Claude Code, VS Code and GitHub. Free for 5,000 requests a month.
Free Options
Launch Team
ArticosYour audience, ready to react to your messaging in minutes. Try free
Promoted
Hey Product Hunt! I'm Ricky, founder of BLAZLE.
AI coding agents now run shell commands, edit repos and call MCP tools on their own. One poisoned file or careless prompt can leak a secret or run something destructive.
BLAZLE inspects every prompt, command, diff and tool call in-flight, in under a millisecond for most checks:
• Redacts secrets before egress
• Blocks prompt injection in 30+ languages
• Holds risky actions for human approval
• Turns your existing security policies into enforceable rules
The free tier covers 5,000 intercepted requests a month, no card needed. Would love your feedback and any bypass you can find!