---
id: gh-convex-sentinel
name: "convex-sentinel"
url: https://skills.yangsir.net/skill/gh-convex-sentinel
author: convex
domain: ai-system-observability-sre
tags: ["observability", "error-tracking", "convex", "monitoring", "sre"]
install_count: 34700
rating: 4.60 (120 reviews)
github: https://github.com/convex/agent-skills/tree/main/skills/convex-sentinel
---

# convex-sentinel

> 该 Skill 可在自有 Convex 部署中接入 Sentinel 生产错误捕获，把服务端函数异常、前端 JS/React 崩溃、OCC 与扩容信号写入部署内的数据表，并在写入时默认脱敏。数据始终留在用户自己的部署里，不经过任何第三方，配合 triage 定时任务还能自动分类并为反复出现的错误生成修复 PR。

**Stats**: 34,700 installs · 4.6/5 (120 reviews)

## Before / After 对比

### 生产错误从发生到定位

**Before**:

生产环境报错后，开发者只能靠用户反馈和零散日志被动发现，翻遍多个日志入口才能拼出一次崩溃的上下文，平均要几十分钟才确认问题真实存在。

**After**:

此技能把服务端异常与前端崩溃统一写入部署内数据表，并附带面包屑上下文，错误一发生即可在 CLI 中读到，定位时间压缩到几分钟。

| Metric | Before | After | Change |
|---|---|---|---|
| 平均发现时长 | 45分钟 | 3分钟 | -93% |
| 首次定位耗时 | 60分钟 | 8分钟 | -87% |
| 需人工比对的日志入口数 | 5个 | 1个 | -80% |

### 错误分类与自愈修复

**Before**:

错误堆在表里没人分类，开发者每周要花数小时人工判断哪些是偶发抖动、哪些必须修，反复出现的同类错误往往拖到一周后才有人处理。

**After**:

该 Skill 的 triage 定时任务自动给每条错误分类，对反复出现的非瞬时错误直接交给 ai-runner 开修复 PR，人工只需审核结果。

| Metric | Before | After | Change |
|---|---|---|---|
| 每周人工分类工时 | 4小时 | 0.5小时 | -88% |
| 反复错误修复周期 | 7天 | 1天 | -86% |
| 分类自动化覆盖率 | 0% | 90% | +90% |

## Readme

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

# Capture production errors in your own deployment

Install `@convex-dev/sentinel` to capture production errors (server function failures, client JS/React crashes, OCC and scale signals) into a table in the user's OWN deployment, redacted at write time, then react to new ones. Data never leaves the user's deployment.

## Workflow

1. Install the component: `app.use(sentinel)` in `convex/convex.config.ts`.
2. Wire the client SDK: a React error boundary plus `window.onerror`/`unhandledrejection` and breadcrumbs.
3. Redaction runs at write time and is on by default (default-deny on secret key names and value patterns).
4. Read recent errors with the Convex CLI (`convex data`, `run-once-query`); react to new ones via the monitor's `prod_error` event.
5. Optionally enable the self-healing cron: `triage` classifies each error and, for recurring non-transient ones, hands it to ai-runner to open a fix PR.

## Rules

- Redaction is mandatory and on by default — never store raw secrets; the agent's reads reach the model provider.
- Data stays in the user's deployment; never send it to a third party.
- Sample and cap to control volume and cost.
- Capturing PROD errors needs a deployed cloud app; install works anonymously.


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