本文へ移動
cccskills
無料GitHub で公開

post-deployment-readonly-verification

Generic read-only post-deployment verification workflow for confirming service health, artifact activation, routing, logs, and rollback signals after a deployment action.

インストール方法を見る

含まれるファイル(1)

  • SKILL.md4.3 KB

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

Post-Deployment Read-Only Verification

Purpose

Verify whether a deployment is actually healthy after a change. The workflow distinguishes "command executed" from "service currently healthy" by checking runtime state, artifact activation, routes, logs, and user-visible behavior without making further changes.

When to use

  • After a deployment, restart, rollback, config change, package replacement, or migration.
  • When a service reports active but user-facing checks still fail.
  • Before telling a user a deployment succeeded.
  • When a previous deployment action was performed by another person or agent and needs independent verification.

Safety rules

  • Default workflow is read-only.
  • Do not restart, reload, edit, redeploy, roll back, clear caches, run migrations, or execute write-path tests without explicit approval.
  • Do not print raw production logs, customer data, order data, tokens, cookies, JWTs, database URLs, internal IPs, or real domains.
  • Replace identifiers with <PROJECT_NAME>, <SERVICE_NAME>, <APP_DOMAIN>, <PROJECT_ROOT>, <SERVER_IP>, and <ARTIFACT_NAME>.
  • Health checks should be safe GET/read-only probes unless the user approves write-path validation.

Read-only workflow

  1. Confirm what changed: artifact, config, service, migration, rollback, or unknown. Record who/what performed the action if known.
  2. Check current service state, not only the state immediately after deployment: active state, substate, main PID, restart count, exit status, and uptime.
  3. Verify artifact activation: running command line, file timestamp, version endpoint if safe, static asset timestamp, container image digest, or application startup banner.
  4. Check local ports and process bindings against expected ports.
  5. Check logs since deployment time: summarize startup success, warnings, errors, crash loops, migration messages, and repeated exceptions without raw sensitive content.
  6. Perform safe local probes: health endpoint, root page, read-only API, static asset, or internal readiness endpoint.
  7. Perform safe external-route probes if allowed: reverse proxy domain, TLS status, CDN/tunnel route, gateway prefix, and status code class. Avoid exposing real domains in public reports.
  8. Validate dependency signals: database pool ready, queue/worker connected, external API readiness, proxy state, and scheduled task handoff if relevant.
  9. Compare observed behavior to rollback criteria. Identify whether rollback should be considered, but do not execute it.
  10. Produce a pass/warn/fail verdict with evidence and next steps.

Evidence to collect

  • Deployment time or change window.
  • Current service active/substate, PID, restart count, uptime.
  • Artifact/version activation evidence.
  • Port and process binding summary.
  • Safe log summary since deployment.
  • Local read-only probe results.
  • External read-only probe results if in scope.
  • Dependency readiness summary.
  • Rollback trigger indicators.

Stop conditions

  • Verification requires write-path tests without approval.
  • Logs contain sensitive data that cannot be summarized safely.
  • The deployment time or changed component is unknown and cannot be inferred.
  • Current checks indicate active data corruption, migration failure, or customer-impacting outage requiring human decision.
  • A rollback or restart appears necessary but has not been approved.

Optional user-approved actions

Only after explicit user approval:

  • Restart/reload service or proxy.
  • Execute rollback.
  • Run write-path smoke tests.
  • Run data migration verification queries requiring elevated access.
  • Patch configuration or redeploy an artifact.
  • Increase logging or add temporary diagnostics.

Final report format

Summary: <pass/warn/fail verdict>
Change verified: <artifact/config/service/unknown>
Current runtime: <service/PID/restarts/uptime>
Artifact activation: <evidence>
Read-only probes: <local/external status summary>
Logs since change: <safe summary>
Dependencies: <summary>
Rollback signals: <none/present/unknown>
Recommended next step: <observe / approval-required fix / rollback decision>
Needs approval: <specific action if any>

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

同じリポジトリのスキル

概要と使いどころ

Route CEO-first multi-window work into owner-led role loops. Use for 总控/CEO、架构/CTO、内容主编、角色派发与复用、模型路由、来源窗口回调、role-windows 台账、skill 命中统计或 multi-window loop engineering。

日本語の概要は準備中です。原文の説明を表示しています。

Dirtytrii/codex-skills112026年10月9日 更新

Read-only first workflow for diagnosing application incidents by collecting service, log, network, configuration, and functional evidence before proposing any fix.

日本語の概要は準備中です。原文の説明を表示しています。

Dirtytrii/codex-skills112026年10月9日 更新

Run user-authorized black-box web security tests and evidence-based reports. Use for 黑盒安全测试、渗透测试、数据泄露、未授权访问、公开 JS/API、登录爆破防护、CORS 或安全响应头检查。

日本語の概要は準備中です。原文の説明を表示しています。

Dirtytrii/codex-skills112026年10月9日 更新

Choose the narrowest reliable browser surface across Codex Browser, signed-in Chrome, Playwright CLI, and platform scripts. Use for browser automation, logged-in workflows, UI inspection, screenshots, form filling, publish previews, or frontend QA.

日本語の概要は準備中です。原文の説明を表示しています。

Dirtytrii/codex-skills112026年10月9日 更新

提议并执行 rubric 或 bucket 升级。两种模式:**完整 rubric bump**(最高风险动作,5 步强制 + 跨模型审核)和 **--bucket-only 轻量重校**(只换 bucket 边界,不动 rubric 公式)。**Phase 2 强制走 cheat-score-blind sub-agent 给校准池重打分**——不接受 self-scored fallback。触发词:"升级 rubric"/"bump rubric"/"更新公式"/"我想加一个维度"/"调整权重"/"重校桶"/"recalibrate bucket"。

日本語の概要は準備中です。原文の説明を表示しています。

Dirtytrii/codex-skills112026年10月9日 更新

初始化 cheat-on-content 内容校准工作区。仅在用户明确开始该工作流,或其子 skill 确认缺少 .cheat-state.json 时使用;不响应一般项目初始化、安装诊断或 Codex 首次会话。

日本語の概要は準備中です。原文の説明を表示しています。

Dirtytrii/codex-skills112026年10月9日 更新

Dirtytrii のスキルをすべて見る

このスキルの問題を報告する