How to Fix OpenClaw’s Feishu API Quota Exhaustion with Self‑Repair

Problem Overview

Many OpenClaw users who connect the AI assistant to Feishu encounter the warning “Dear developer, your enterprise API call quota is about to be exhausted.” This happens even for users who make only occasional requests because the free personal Feishu plan limits the number of API calls per month.

The core issue is that OpenClaw’s Feishu integration calls the /open-apis/bot/v3/info endpoint every minute. For a single agent this amounts to:

1 call per minute × 1440 minutes per day = 1440 calls per day

1440 calls per day × 30 days ≈ 43,200 calls per month

These calls alone consume the entire free quota, and additional checks add further usage.

Root‑Cause Analysis

The excessive calls originate from three places in the codebase:

Monitor startup – extensions/feishu/src/monitor.ts invokes fetchBotOpenId() → probeFeishu() each time the Feishu account starts.

Channel status check – extensions/feishu/src/channel.ts calls probeAccount() → probeFeishu() during each channel health check.

Gateway health check – Configured via gateway.channelHealthCheckMinutes (default 30 min), which triggers the same probe regularly.

Solution 1 – Add a Cache Layer

The AI assistant generated an optimized probe.ts (named probe_optimized.ts) that introduces a 5‑minute TTL cache to store the result of probeFeishu(). The cache key is appId:appSecret, and the implementation automatically evicts entries when more than ten are stored.

import { createFeishuClient, type FeishuClientCredentials } from "./client.js";
import type { FeishuProbeResult } from "./types.js";
// Cache for bot probe results with TTL (5 minutes)
const botProbeCache = new Map<string, { result: FeishuProbeResult; timestamp: number }>();
const CACHE_TTL_MS = 5 * 60 * 1000; // 5 minutes cache
export async function probeFeishu(creds?: FeishuClientCredentials): Promise<FeishuProbeResult> {
  if (!creds?.appId || !creds?.appSecret) {
    return { ok: false, error: "missing credentials (appId, appSecret)" };
  }
  const cacheKey = `${creds.appId}:${creds.appSecret}`;
  const cached = botProbeCache.get(cacheKey);
  if (cached && Date.now() - cached.timestamp < CACHE_TTL_MS) {
    return cached.result;
  }
  try {
    const client = createFeishuClient(creds);
    const response = await (client as any).request({ method: "GET", url: "/open-apis/bot/v3/info", data: {} });
    if (response.code !== 0) {
      return { ok: false, appId: creds.appId, error: `API error: ${response.msg || `code ${response.code}`}` };
    }
    const bot = response.bot || response.data?.bot;
    const result: FeishuProbeResult = { ok: true, appId: creds.appId, botName: bot?.bot_name, botOpenId: bot?.open_id };
    botProbeCache.set(cacheKey, { result, timestamp: Date.now() });
    if (botProbeCache.size > 10) {
      const oldestKey = Array.from(botProbeCache.entries())
        .sort((a, b) => a[1].timestamp - b[1].timestamp)[0]?.[0];
      if (oldestKey) botProbeCache.delete(oldestKey);
    }
    return result;
  } catch (err) {
    return { ok: false, appId: creds.appId, error: err instanceof Error ? err.message : String(err) };
  }
}

Replacing the original extensions/feishu/src/probe.ts with this file eliminates redundant API calls within the cache window.

Solution 2 – Reduce Health‑Check Frequency

Modify openclaw.json to increase the health‑check interval from 30 minutes to 60 minutes:

{
  "gateway": {
    "channelHealthCheckMinutes": 60 // changed from 30
  }
}

Solution 3 – Optimize Monitor Startup Logic

Adjust monitor.ts to add a guard that prevents the probe from running on repeated starts. The AI assistant supplied a screenshot of the revised logic (omitted here for brevity).

Implementation Artifacts

The assistant also produced four supporting files: feishu_api_optimization.md – detailed technical analysis and full optimization plan. apply_feishu_optimization.sh – automation script to apply the changes. probe_optimized.ts – the cached probe implementation shown above. QUICK_FIX.md – quick‑start guide.

Before vs. After

Before : Every agent start and every health check triggers a fresh /open-apis/bot/v3/info request.

After : Calls within a 5‑minute window hit the cache, reducing API usage by more than 90 %.

Results

Post‑optimization logs (shown in the article’s screenshots) confirm a dramatic drop in request frequency. Users also note that Feishu’s free tier quota may have been reduced to 10 k calls per month, and a link to apply for higher limits is provided.

Conclusion

By adding a lightweight cache, extending health‑check intervals, and tightening monitor startup logic, OpenClaw’s Feishu integration can operate within the free‑plan quota, avoiding service interruptions without requiring external tooling.