Home › Documentation › Validation & robustness
Artificial Intelligence

AI Analysis

AniQuant connects to the Claude API (Anthropic) to interpret your studies. It generates no signals and places no trades: it reads the metrics you have already computed and returns a structured critical assessment.

It needs your API key: Settings → AI tab. While it answers, a progress dialog with a clock keeps you informed. Model: claude-opus-5.

🔎

Analyze with AI

Right-click in Performance, Walk-Forward, PBO/DSR or Survival. It sends the strategy plus its metrics and returns a verdict:

recommendation confidence overfit risk robustness OOS quality

Plus strengths, risks, signs of overfitting and ideas for improvement. An HTML report you can save in C:\AniQuant\Informes\.

✨

Improve with AI

From the AQL code editor. The AI proposes 3-4 improved variants in the AQL language towards a goal (robustness, drawdown, a regime filter…).

Every proposal arrives already backtested — AniQuant's engine is the judge, not the AI — with its metrics against the baseline. You choose Load or Test.

⚠️
Guidance, not advice

The analysis is a critical second opinion to help you discard and improve. The decision to trade for real is always yours.

Try it yourself

AniQuant can be tried free for 14 days, with every module and no card. If you have the book, 30.

← Previous
360 Analysis
Next →
Scatter chart: every result as a point
More in Validation & robustness