🪢 Open source agent evals & observability: Trace, evaluate, and improve LLM applications with one open platform.
TypeScript analyticsautogenevaluationlangchainlarge-language-modelsllama-indexllmllm-evaluationllm-observabilityllmopsmonitoringobservabilityopen-sourceopenaiplaygroundprompt-engineeringprompt-managementself-hostedycombinator
import { toCompactVerbosityChatML } from "./chatML/toCompactVerbosityChatML";
/**
* Returns a compact representation of IO data for display in tables.
* Strategy: Try ChatML extraction if possible.
*
* @param io - The input or output data to compact
* @returns Compact representation or null if no data
*/
export function toCompactVerbosity(io: unknown): {
success: boolean;
data: string | null;
} {
if (io === undefined || io === null) return { success: false, data: null };
// Parse stringified JSON if needed
let parsedIO = io;
if (typeof io === "string") {
try {
parsedIO = JSON.parse(io);
} catch {
// Continue with original input
}
}
// Try ChatML compact representation extraction
const chatMLCompact = toCompactVerbosityChatML(parsedIO);
if (chatMLCompact.success) {
return { success: true, data: chatMLCompact.data };
}
return { success: false, data: null };
}