skills.rest
buzzbysolcex
buzzbysolcexCOMMUNITY
Explore all skills by this author →
Website𝕏@BuzzBySolCexGitHub

MiroFish Dual-Brain (90% Opus / 10% Ollama)

Simulate token listing probability with a 500-agent hybrid market model and AMM price convergence.

5|2|Updated Feb 6, 2026
Search Tags:#Ollama#AMM#LLM routing#market simulation#token scoring#listing probability#EV calculation
One-click install
npx skills add https://github.com/buzzbysolcex/buzz-bd-agent --skill mirofish-dual-brain-90-opus-10-ollama
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: MiroFish Dual-Brain (90% Opus / 10% Ollama)
Source: https://github.com/buzzbysolcex/buzz-bd-agent/tree/main/.claude/skills/mirofish-simulation
Command: npx skills add https://github.com/buzzbysolcex/buzz-bd-agent --skill mirofish-dual-brain-90-opus-10-ollama

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It removes guesswork in token listing probability by simulating market behavior with large-scale agent trading rather than relying on single-pass heuristics.

Core Features & Use Cases

  • 500-agent hybrid simulation: Runs 30 LLM-driven agents (Ollama for early degen rounds, Claude Opus otherwise) plus 470 heuristic agents to model consensus and market dynamics.
  • 3-round AMM-based market convergence: Uses a constant-product AMM to translate agent trades into a final listing probability score.
  • Evidence-grade scoring metrics: Outputs microbuzz score, EV, round-by-round price evolution, LLM consensus strength, and trade volume as explainable signals.

Quick Start

Run a 3-round 500-agent simulation for a token address and return the final microbuzz listing probability and EV score.

Frequently Asked Questions about MiroFish Dual-Brain (90% Opus / 10% Ollama)

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I calculate token listing probability using market simulation instead of single-pass heuristics?▼

Token listing probability is calculated by running a 500-agent hybrid market simulation that models consensus and market dynamics through 3 rounds of AMM price convergence, replacing single-pass heuristics with large-scale agent trading behavior.

How does AMM price convergence work for token scoring in a multi-round simulation?▼

AMM price convergence for token scoring uses a constant-product automated market maker to translate 500 agents' trades into a final microbuzz listing probability score across 3 rounds, persisting round prices and EV into a SQLite schema.

Can I use Ollama and Claude Opus together for LLM routing in market simulations?▼

Yes, LLM routing between Ollama and Claude Opus is supported by running 30 LLM-driven agents, using Ollama for early degen rounds and Claude Opus for subsequent rounds, alongside 470 heuristic agents for market modeling.

What explainable signals does a 500-agent token scoring simulation output?▼

A 500-agent token scoring simulation outputs microbuzz score, expected value (EV), round-by-round price evolution, LLM consensus strength, and trade volume as evidence-grade, explainable signals for listing likelihood.

Do I need cross-chain context data to estimate token listing likelihood accurately?▼

Cross-chain context data, along with token and social data, is required to estimate listing likelihood accurately across multiple simulation rounds, providing the necessary inputs for the 500-agent hybrid market model.

Related Skills

Explore More Data & Analytics Skills→
readtheskillreadtheskill
3

token-metadata

Fetch SPL token names, symbols, logos, and social links via Solana Agent Kit endpoints.

Official
Intermediate
jkfeejkfee

research-review

Automate multi-round external research critique via Codex MCP.

Community
Advanced
nirholasnirholas
113

pump-bonding-curve

Implement constant-product bonding curve math for Pump token pricing and trading.

Community
Advanced
arbazkhan971arbazkhan971
26

tokens

Measure per-round token usage in godmode loops and log to a TSV file.

Community
Advanced