---
id: gh-convex-monitor
name: "convex-monitor"
url: https://skills.yangsir.net/skill/gh-convex-monitor
author: convex
domain: ai-system-observability-sre
tags: ["convex", "observability", "error-monitoring", "event-driven", "sre"]
install_count: 34800
rating: 4.60 (120 reviews)
github: https://github.com/convex/agent-skills/tree/main/skills/convex-monitor
---

# convex-monitor

> 监听 Convex 应用的下一个开发或生产错误/请求事件，一旦触发即返回类型化事件供修复。它把轮询式排查改为阻塞等待，竞速本地错误日志、部署订阅与 Sentinel 生产错误行，开发者可第一时间响应而不是被动翻日志。

**Stats**: 34,800 installs · 4.6/5 (120 reviews)

## Before / After 对比

### 错误发现与响应速度对比

**Before**:

开发者只能反复手动刷新日志、轮询部署状态，错误发生后往往要等几分钟甚至更久才知道，排查上下文还得自己拼凑。

**After**:

此技能阻塞等待下一个类型化事件，本地错误日志、部署订阅与生产错误行竞速返回首个触发事件，发现即定位、定位即修复。

| Metric | Before | After | Change |
|---|---|---|---|
| 错误发现延迟 | 25分钟 | 1分钟 | -96% |
| 无效轮询次数 | 60次/小时 | 1次/小时 | -98% |
| 上下文拼凑耗时 | 15分钟 | 2分钟 | -87% |

## Readme

<!-- GENERATED from convex-agents content/capabilities/monitor.json — do not edit by hand. -->

# Watch for the next thing to react to

Block on the next typed event instead of polling. Races local error logs, deployment subscriptions, and Sentinel prod-error rows; returns the first to fire (or a quiet heartbeat).

## Workflow

1. Call `wait_for_event` with {project_dir, event_kinds, timeout_ms}.
2. On kind=convex_error/next_error: decode and fix it. On kind=prod_error: triage (see sentinel) and fix. On kind=feature_request: build it. On kind=quiet: loop.
3. Where a harness has no blocking MCP (e.g. Copilot cloud), the pack runs a poll loop with the SAME event contract — same behavior, different mechanism.

## Rules

- Prefer the blocking tool; fall back to a poll loop only where blocking MCP is weak.
- The event schema is fixed and versioned — the same trigger yields the same typed event.
- Prod events (kind=prod_error) require a deployed cloud app plus Sentinel.


---
*Source: https://skills.yangsir.net/skill/gh-convex-monitor*
*Markdown mirror: https://skills.yangsir.net/api/skill/gh-convex-monitor/markdown*