Bug type

Regression (worked before, now fails)

Summary

Cron job shows contradictory state: lastDelivered=true (message sent successfully) but lastRunStatus=error (marked as failed), causing false positive error alerts and accumulated consecutiveErrors despite successful delivery.

Steps to reproduce

1. Create a Cron job with delivery.mode: "announce" to Feishu channel
2. Let the job execute (or manually trigger with openclaw cron run --id <id>)
3. Observe the job successfully sends message to Feishu group
4. Run openclaw cron list --include-disabled
5. Check job state: lastDelivered=true but lastRunStatus=error

Expected behavior

When lastDelivered=true (message successfully delivered), the lastRunStatus should be "ok" and consecutiveErrors should be reset to 0. The state fields should be consistent: successful delivery = successful execution.

Actual behavior

Job state shows contradictory values:

• state.lastDelivered: true (message was sent successfully to Feishu)
• state.lastRunStatus: "error" (but marked as failed)
• state.consecutiveErrors: keeps incrementing (4+ errors despite successful delivery)

Additional issue: Found hardcoded platform-specific error messages in code:

throw new Error("telegram sendMessage failed: ...");

This should be parameterized since the actual platform is Feishu, not Telegram.

OpenClaw version

2026.3.2

Operating system

Windows 11 (Windows_NT 10.0.26200)

Install method

npm global

Model

bailian/qwen3.5-plus

Provider / routing chain

openclaw -> bailian (DashScope) -> qwen3.5-plus

Config file / key location

No response

Additional provider/model setup details

No response

Logs, screenshots, and evidence

Impact and severity

Affected: All users using Cron jobs for monitoring and scheduled tasks
Severity: Medium-High (causes false positive alerts, confusing monitoring dashboards)
Frequency: 100% reproducible when Cron job successfully delivers messages
Consequence:

• Monitoring reports show errors when jobs actually succeeded
• consecutiveErrors accumulates incorrectly, may trigger false alerts
• Users cannot trust the job state for health monitoring
• Hardcoded error messages cause confusion during debugging

Additional information

Root Cause Analysis:

The state update logic in Cron job execution doesn't properly correlate delivery success with execution status. When a job successfully delivers its output, the lastRunStatus should reflect that success.

Suggested Fix:

In the Cron job state update logic:

if (job.state.lastDelivered === true) {
    job.state.lastRunStatus = "ok";
    job.state.consecutiveErrors = 0;  // Reset error count
}

Files Involved:

• packages/plugin-sdk/src/cron/types.ts - Type definitions
• dist/plugin-sdk/cron/types.d.ts - Shows the state fields are meant to be independent
• dist/pi-embedded-*.js - Execution logic (compiled)

Additional Bug: Hardcoded Error Messages

Found in dist/pi-embedded-CtM2Mrrj.js:76882:

throw new Error("telegram sendMessage failed: ...");

This should be parameterized to support multiple platforms (Feishu, Discord, Telegram, etc.).