Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add nongjun/feishu-cursor-claw --skill agent-self-verificationgit clone --depth 1 https://github.com/nongjun/feishu-cursor-clawWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/nongjun/feishu-cursor-claw/agent-self-verification)<a href="https://agentmods.dev/skills/nongjun/feishu-cursor-claw/agent-self-verification"><img src="https://agentmods.dev/badge/skills/nongjun/feishu-cursor-claw/agent-self-verification/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/nongjun/feishu-cursor-claw/agent-self-verification"><img src="https://agentmods.dev/badge/skills/nongjun/feishu-cursor-claw/agent-self-verification.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00097 | $0.01834 |
| Opus 5 | $0.00048 | $0.00917 |
| Sonnet 5 | $0.00019 | $0.00367 |
| Haiku 4.5 | $0.00010 | $0.00183 |
Grade A, and why
agent-self-verification scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -s -o /dev/null -w "%{http_code}" https://<domain>/api/health How it starts
The opening of the file, as written. The whole thing — 215 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent 闭环自验证工作流
核心理念:写完代码不是终点,验证通过才算完成。 Agent 应像人类工程师一样,改完代码后"打开浏览器看看对不对、查看日志有没有报错、确认性能正常"。
何时触发
每次完成以下任何改动后,必须进入验证循环:
- 前端 UI 变更(组件、样式、布局)
- 后端 API 变更(接口、逻辑、数据结构)
- Bug 修复
- 配置变更(Docker、Nginx、环境变量)
- 数据库 schema 变更
验证循环流程
┌─────────┐
│ 写代码 │
└────┬────┘
▼
┌─────────┐
│ 构建/部署 │ ← docker compose up --build / npm run build
└────┬────┘
▼
┌─────────────┐
│ 静态检查 │ ← Linter + TypeCheck
└────┬────────┘
▼
┌─────────────┐
│ UI 验证(眼睛) │ ← 浏览器快照 + 截图 + 控制台
└────┬────────┘
▼
┌──────────────────┐
│ 系统验证(听诊器) │ ← 容器日志 + 健康检查 + 网络请求
└────┬─────────────┘
▼
┌─────────┐ 失败 ┌─────────┐
│ 判定结果 │─────────→│ 定位修复 │──→ 返回"写代码"
└────┬────┘ └─────────┘
│ 通过
▼
┌─────────┐
│ 任务完成 │
└─────────┘
第一步:构建与部署验证
代码改动后,先确认构建成功:
# 前端项目
docker compose up --build -d <service-name>
docker logs <container-name> --tail 30
# 检查容器状态
docker ps --filter "name=<container-name>" --format "table {{.Names}}\t{{.Status}}\t{{.Ports}}"
判定标准:容器状态为 Up,无 Restarting,日志无 Error/Exception。
第二步:静态检查
# 使用 ReadLints 工具检查已编辑文件
# 如果有 TypeScript,检查类型错误
docker exec <container-name> npx tsc --noEmit 2>&1 | tail -20
判定标准:无新增 linter 错误,无类型错误。
第三步:UI 验证(装"眼睛")
对于前端变更,使用浏览器 MCP 工具验证:
3.1 导航到目标页面
browser_navigate → 目标 URL
browser_snapshot → 获取页面结构(相当于 DOM "X光片")
3.2 验证页面元素
browser_snapshot(interactive: true) → 检查交互元素是否存在且可操作
browser_is_visible(selector) → 确认关键元素可见
browser_get_input_value → 验证表单默认值
3.3 验证交互行为
browser_click / browser_fill → 模拟用户操作
browser_snapshot(includeDiff: true) → 对比操作前后的页面变化
3.4 检查前端错误
browser_console_messages → 检查是否有 JS 错误或警告
browser_network_requests → 检查 API 请求是否成功(无 4xx/5xx)
3.5 视觉确认
browser_take_screenshot(fullPage: true) → 截取完整页面
判定标准:
- 目标元素存在且可见
- 控制台无 error 级别消息
- 网络请求全部成功
- 交互行为符合预期
第四步:系统验证(装"听诊器")
4.1 容器日志检查
# 查看最近日志,过滤错误
docker logs <container-name> --since 2m 2>&1 | grep -iE "error|exception|traceback|failed"
# 后端 API 日志
docker logs <backend-container> --since 2m 2>&1 | tail -50
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 9d ago First seen · 215 lines · 97 tokens per session scan A b8be740fcc24
agent-self-verification is a skill published in the GitHub repository nongjun/feishu-cursor-claw (14 stars, last pushed 1mo ago), licensed MIT. It adds 97 tokens to every session and 1,834 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
firebase-crashlytics
Use when implementing crash reporting, capturing fatal/non-fatal errors, recording isolate/async exceptions, customizing reports, or uploading obfuscated symbols.
flutter-errors
Use when hitting layout errors (RenderFlex overflow, unbounded constraints, RenderBox not laid out), scroll errors, or setState-during-build errors.
audit-performance
Audit runtime performance (CWV, load priority). Use when "slow page", "LCP/INP/CLS", "fetchpriority", "early hints", "speculation rules", "bfcache", or "long tasks". JS payload → audit-bundle-size. Instant nav implement → enhance-web-instant-nav. Breaking point → test-load.
debug-fe-be-integration
Diagnose and fix frontend↔backend contract failures by tracing client requests, server logs, validation, auth, and responses on both sides. Use for API 4xx/5xx, mismatched payloads, validation failures, or "frontend and backend disagree". Single-layer errors → debug-error.
debug-sentry-monitor
Operate Sentry: triage/fix unresolved issues, reduce noise, audit instrumentation, and monitor after deploy. Use when "check Sentry", "fix Sentry issues", "review production errors", or "improve error tracking". One named bug through PR → workflow-fix-and-ship. Plan-only observability → plan-error-handling.
backend-db-performance
Optimize slow queries, indexes, and N+1s. Use when "slow query", "database performance", "add an index", or "N+1". Schema consistency → audit-db-schema. RLS access control → plan-rls-audit.