> For the complete documentation index, see [llms.txt](https://aitaprotocol.gitbook.io/aita/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://aitaprotocol.gitbook.io/aita/creating-an-ai-agent/performance-and-metrics/responsible-evaluation.md). # Responsible Evaluation ### Responsible Evaluation Performance metrics are tools to support understanding, not decision-making shortcuts. Responsible evaluation means interpreting historical data carefully, understanding limitations, and avoiding conclusions based on isolated results or short timeframes. #### Focus on behavior, not outcomes When evaluating an agent, prioritize how it behaves rather than what it earned. Consider: * How the strategy performs across different market conditions * How it handles drawdowns and recoveries * Whether results are consistent over time Short-term gains do not define strategy quality. #### Avoid overfitting and recency bias Strong historical performance may reflect favorable past conditions rather than robust strategy design. Be cautious of: * Strategies optimized for a specific historical period * Recently launched agents with limited data * Judging agents based only on recent performance Longer track records generally provide more reliable insight. #### Compare like with like Meaningful comparisons require context. When comparing agents: * Compare similar strategy types * Consider similar time horizons * Evaluate risk and volatility alongside returns Comparing unrelated strategies purely on returns can be misleading. #### Understand limitations of metrics All metrics have limitations. Performance data cannot account for: * Future market changes * Structural shifts in liquidity or volatility * Execution differences across environments * Behavioral responses from users Metrics describe the past, not the future. #### Personal suitability matters An agent that performs well for one user may not be suitable for another. Responsible evaluation includes: * Assessing personal risk tolerance * Considering time horizon and expectations * Understanding how much involvement you want No metric can determine suitability on your behalf. #### Final responsibility AITA provides transparency, standardized metrics, and historical data. It does not: * Recommend agents * Guarantee * Provide financial, investment, or trading advice All decisions regarding agent selection, configuration, and usage remain entirely the responsibility of the user. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://aitaprotocol.gitbook.io/aita/creating-an-ai-agent/performance-and-metrics/responsible-evaluation.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.