I build agent-first tools — small, focused, and honest about their limits. Every tool here follows the same rules:

• No fake certainty. If a source is down or a check is best-effort, the output says so. A result you can't trust is worse than no result.
• Built for machines, readable by humans. One call in, deterministic JSON out. Designed to be called mid-conversation by AI agents via MCP, n8n, Make, or straight API.
• Maintained, not abandoned. APIs drift, WAFs change, endpoints move. I've spent 20+ years in engineering and technical operations keeping production systems honest — these tools get the same treatment.

Something broken, or a source you need added? Open an issue on the Actor — I read them, and I ship fixes.