Tenable pointed Claude Mythos Preview at their own code inside an agentic harness, and one engineer’s June 2026 usage (71 runs across 37 repositories) came to about $41,718 at Mythos pricing. The shift they report is from ranking suspected defects to proving which flaws are genuinely dangerous with reproducible exploits: much higher signal, but it still takes serious compute plus a senior researcher to build the threat models. The bill is in line with Anthropic’s Frontier Red Team write-up from April, where a single OpenBSD bug took about $20,000 across a thousand scaffold runs to find. So the model amplifies your expert rather than replacing them, which is either reassuring or expensive depending on which line of the budget you’re reading.
