Unless there's been more information since their original post (https://red.anthropic.com/2026/mythos-preview/), this is a misleading description of the scaffold. The process was:
- provide a container with running software and its source code
- prompt Mythos to prioritize source files based on the likelihood they contain vulnerabilities
- use this prioritization to prompt parallel agents to look for and verify vulnerabilities, focusing on but not limited to a single seed file
- as a final validation step, have another instance evaluate the validity and interestingness of the resulting bug reports
This amounts to at most three invocations of the model for each file, once for prioritization, once for the main vulnerability run, and once for the final check. The prompts only became more specific as a result of information the model itself produced, not any external process injecting additional information.