Never stop coding. Free MIT AI gateway: one endpoint, 375 providers (150+ free), 1,400+ models Kimi, Claude, GPT, Gemini, GLM, DeepSeek, MiniMax. Works with Claude Code, Codex, Cursor, OpenCode, Cline & Copilot. Quota-aware auto-fallback, RTK+Caveman compression saves 15-95% tokens, MCP/A2A, Desktop/PWA. Built by 900+ contributors
TypeScript a2aai-agentsai-gatewayanthropicclaudeclaude-codeclinecodexcopilotcursordeepseekfree-aigeminikimillm-gatewaymcpopenaiopenai-proxyqwentoken-saver
import { NextRequest, NextResponse } from "next/server";
export const dynamic = "force-dynamic";
import {
getCacheStats,
clearCache,
invalidateByModel,
invalidateBySignature,
invalidateStale,
} from "@/lib/semanticCache";
import { getIdempotencyStats } from "@/lib/idempotencyLayer";
import { getCacheMetrics, getCacheTrend } from "@/lib/db/settings";
import { getCachedSettings } from "@/lib/db/readCache";
import {
getCacheMetricsMinComputeMs,
invalidateAnalyticsCachePrefix,
serveAnalyticsCached,
} from "@/lib/usage/analyticsResponseCache";
import { isTrustedLoopbackInternalServiceRequest } from "@/lib/api/internalServiceAuth";
import { isAuthenticated, isCacheScopedKey } from "@/shared/utils/apiAuth";
import { CACHE_READ_SCOPE, CACHE_WRITE_SCOPE } from "@/shared/constants/managementScopes";
import { sanitizeErrorMessage } from "@omniroute/open-sse/utils/error";
/** Key prefix of the opt-in history entries in the shared analytics response cache. */
const CACHE_METRICS_KEY_PREFIX = "cache-metrics?";
function errorMessage(error: unknown): string {
return sanitizeErrorMessage(error);
}
export async function GET(req: NextRequest) {
if (
!isTrustedLoopbackInternalServiceRequest(req) &&
!(await isAuthenticated(req)) &&
!(await isCacheScopedKey(req, CACHE_READ_SCOPE))
) {
return NextResponse.json({ error: "Unauthorized" }, { status: 401 });
}
try {
const { searchParams } = new URL(req.url);
const rawHours = parseInt(searchParams.get("trendHours") || "24", 10);
const trendHours = Math.min(720, Math.max(1, Number.isNaN(rawHours) ? 24 : rawHours));
// Opt-in (OMNIROUTE_CACHE_METRICS_MIN_COMPUTE_MS): serve the usage-history
// part from the shared response cache. Unset = every read is live, as before.
const historyMinComputeMs = getCacheMetricsMinComputeMs();
const history =
historyMinComputeMs === null
? null
: await readCachedHistory(trendHours, historyMinComputeMs);
const cacheStats = getCacheStats();
// Promise.all allows concurrent execution; if one rejects, others still resolve,
// which is safe for this read-only stats endpoint.
const [idempotencyStats, promptCacheMetrics, trend, settings] = await Promise.all([
getIdempotencyStats(),
history ? history.body.promptCache : getCacheMetrics(),
history ? history.body.trend : getCacheTrend(trendHours),
getCachedSettings().catch(() => ({})),
]);
const response = NextResponse.json({
semanticCache: cacheStats,
promptCache: promptCacheMetrics,
trend,
idempotency: idempotencyStats,
config: {
semanticCacheEnabled: settings.semanticCacheEnabled !== false,
},
});
// hit/miss describes the cached history part only; fresh counters are always live.
if (history?.cacheState) response.headers.set("x-analytics-cache", history.cacheState);
return response;
} catch (error) {
return NextResponse.json({ error: errorMessage(error) }, { status: 500 });
}
}
export async function DELETE(req: NextRequest) {
if (
!isTrustedLoopbackInternalServiceRequest(req) &&
!(await isAuthenticated(req)) &&
!(await isCacheScopedKey(req, CACHE_WRITE_SCOPE))
) {
return NextResponse.json({ error: "Unauthorized" }, { status: 401 });
}
try {
const { searchParams } = new URL(req.url);
const model = searchParams.get("model");
const signature = searchParams.get("signature");
const staleMsParam = searchParams.get("staleMs");
// Enforce mutual exclusivity — only one invalidation mode per request
const paramCount = [model, signature, staleMsParam].filter(Boolean).length;
if (paramCount > 1) {
return NextResponse.json(
{
error:
"Only one invalidation parameter (model, signature, or staleMs) may be provided per request.",
},
{ status: 400 }
);
}
if (model) {
const removed = invalidateByModel(model);
invalidateAnalyticsCachePrefix(CACHE_METRICS_KEY_PREFIX);
return NextResponse.json({ ok: true, invalidated: removed, scope: "model", model });
}
if (signature) {
const removed = invalidateBySignature(signature);
invalidateAnalyticsCachePrefix(CACHE_METRICS_KEY_PREFIX);
return NextResponse.json({ ok: true, invalidated: removed ? 1 : 0, scope: "signature" });
}
if (staleMsParam) {
const maxAgeMs = parseInt(staleMsParam, 10);
if (Number.isNaN(maxAgeMs) || maxAgeMs <= 0) {
return NextResponse.json(
{ error: "staleMs must be a positive integer (milliseconds)." },
{ status: 400 }
);
}
const removed = invalidateStale(maxAgeMs);
invalidateAnalyticsCachePrefix(CACHE_METRICS_KEY_PREFIX);
return NextResponse.json({ ok: true, invalidated: removed, scope: "stale", maxAgeMs });
}
// Full clear
const cleared = clearCache();
invalidateAnalyticsCachePrefix(CACHE_METRICS_KEY_PREFIX);
return NextResponse.json({ ok: true, cleared, scope: "all" });
} catch (error) {
return NextResponse.json({ error: errorMessage(error) }, { status: 500 });
}
}
/** History part of GET (prompt-cache totals + trend), served through the shared cache. */
async function readCachedHistory(
trendHours: number,
minComputeMs: number
): Promise<{ body: { promptCache: unknown; trend: unknown }; cacheState: string | null }> {
const historyResponse = await serveAnalyticsCached(
`${CACHE_METRICS_KEY_PREFIX}trendHours=${trendHours}`,
async () => {
const [promptCache, trend] = await Promise.all([
getCacheMetrics(),
getCacheTrend(trendHours),
]);
return NextResponse.json({ promptCache, trend });
},
{ minComputeMs }
);
return {
body: (await historyResponse.json()) as { promptCache: unknown; trend: unknown },
cacheState: historyResponse.headers.get("x-analytics-cache"),
};
}