lucemia

Use when creating matplotlib/strategy charts. Use when user asks to "draw", "chart", "plot", or "visualize" a strategy, backtest result, or price series.

Use when optimizing a strategy, model, or system through parallel experimentation — when the search space is large, multiple hypotheses exist, and there is an automated scoring function to evaluate variants.

Use when converting autoresearch markdown results (verified_insights.md and AGENT_R*_RESULTS.md) into structured JSON for reporting or slides generation.

Use when generating a structured markdown report from an autoresearch_result.json file for a given ticker. Triggered when user asks for a report, summary, or writeup of autoresearch results.

Guide for using typed-ffmpeg, a modern Python FFmpeg wrapper with extensive typing support and comprehensive filter support. Use this when working with FFmpeg operations, video/audio processing, or filter graphs in Python.

Multi-agent trading analysis for a stock ticker. Runs technical, news, fundamentals, and macro analysts in parallel, then an adversarial bull/bear debate, then Research Manager, Trader, and Portfolio Manager to produce a final BUY/SELL/HOLD decision with entry, stop, and sizing.