🪢 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
type FileReference = { path: string; sha256Hash: string | null };
export function compareSkillFiles(
before: FileReference[],
after: FileReference[],
) {
const oldFiles = new Map(before.map((file) => [file.path, file]));
const newFiles = new Map(after.map((file) => [file.path, file]));
return [...new Set([...oldFiles.keys(), ...newFiles.keys()])]
.map((path) => {
const oldFile = oldFiles.get(path);
const newFile = newFiles.get(path);
let status: "Added" | "Removed" | "Modified" | "Unchanged" = "Modified";
if (!oldFile) status = "Added";
else if (!newFile) status = "Removed";
else if (
oldFile.sha256Hash !== null &&
oldFile.sha256Hash === newFile.sha256Hash
)
status = "Unchanged";
return { path, oldFile, newFile, status };
})
.sort((a, b) => {
const unchanged =
Number(a.status === "Unchanged") - Number(b.status === "Unchanged");
if (unchanged) return unchanged;
if (a.path === "SKILL.md") return -1;
if (b.path === "SKILL.md") return 1;
return a.path.localeCompare(b.path);
});
}