MiroFish
Tagline
AI simulation chat that turns a question into graph building, multi-agent simulation, and a follow-up prediction report.
Description
MiroFish is an AI simulation chat tool for scenario prediction. It is built around a text-first conversation: you ask a question directly, the way you would talk to a chat assistant, and the system handles seed material, simulation, and reporting as one continuous prediction workflow. Supporting files are optional rather than mandatory, so a session can begin with nothing more than a plain-language question and expand later when grounding is needed. The product accepts text, PDF, Markdown, and TXT input, and its stated purpose is to turn that material into graph building, simulation, reporting, and follow-up chat inside a single AI prediction workflow.
Highlights
- Text-first prompts with optional PDF, Markdown, and TXT attachments
- Automated graph, simulation, and reporting workflow
- Multi-agent personas interacting across social surfaces over multiple rounds
- Result cards with report summary and follow-up chat
- Prediction reports with risk signals, narrative paths, and follow-up questions
Quick facts
Discussion (1)
- dengkui yang
For comparing scenarios, can the report export include the source evidence and agent assumptions alongside the conclusions? That would help readers distinguish a changed premise from variation between simulation runs, especially when discussing risk signals.


