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 konglong87/superPM --skill pm-releasegit clone --depth 1 https://github.com/konglong87/superPMWrote 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/konglong87/superpm/pm-release)<a href="https://agentmods.dev/skills/konglong87/superpm/pm-release"><img src="https://agentmods.dev/badge/skills/konglong87/superpm/pm-release/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/konglong87/superpm/pm-release"><img src="https://agentmods.dev/badge/skills/konglong87/superpm/pm-release.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00060 | $0.05061 |
| Opus 5.5 | $0.00024 | $0.02024 |
| Sonnet 5.5 | $0.00012 | $0.01012 |
| Haiku 4.5 | $0.00006 | $0.00506 |
Grade A, and why
pm-release 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 1mo 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 -I https://app.example.com/health How it starts
The opening of the file, as written. The whole thing — 673 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Preamble
echo "🚀 上线执行方案制定工具已启动"
写入时边界
只讨论时不写文件,启动检查不创建目录。用户要求保存产物时,先确认当前产品、精确输出路径及已有文件的覆盖/追加选择;确认后仅在实际 Write 前创建该输出文件的父目录。完整安装可参见 skills/_shared/artifact-policy.md,单独安装本 skill 也按本段执行。
跨 Agent 交互规则
当流程要求与用户交互时:
- 如果当前环境支持 AskUserQuestion,使用 AskUserQuestion(最佳体验)。
- 如果当前环境不支持 AskUserQuestion,必须用普通聊天消息提出同样问题。
- 一次只问一个问题。
- 提问后必须停止当前回合,等待用户回答(STOP and WAIT)。
- 不得在用户回答前生成文档、写入 docs。
- 已有 docs 文件不能替代本轮用户回答。
执行流程
步骤 1: 确定上线策略
使用 AskUserQuestion:
📦 上线策略选择
选择本次上线的方式:
A) 全量发布(一次性上线所有用户) B) 灰度发布(逐步放开用户比例) C) 蓝绿部署(新旧版本并行) D) 金丝雀发布(小流量验证)
继续询问:
🌍 环境部署策略
需要在哪些环境部署?
A) 仅生产环境 B) 测试环境 → 生产环境 C) 开发环境 → 测试环境 → 预发布环境 → 生产环境 D) 自定义环境流程
步骤 2: 制定上线检查清单
使用 AskUserQuestion:
✅ 上线检查项
必须检查哪些项目?(可多选)
A) 功能测试(核心功能验证) B) 性能测试(压力测试、容量验证) C) 安全检查(漏洞扫描、权限验证) D) 兼容性测试(多端、多浏览器) E) 数据备份(数据库、配置文件) F) 监控告警(日志、指标、告警规则) G) 文档完备(用户手册、运维文档) H) 全部检查
步骤 3: 规划发布时间
使用 AskUserQuestion:
⏰ 发布时间窗口
选择合适的发布时间:
A) 工作日白天(便于快速响应问题) B) 工作日夜间(用户量少,影响小) C) 周末夜间(最低峰时段) D) 根据业务特点灵活选择
继续询问:
📅 发布节奏
发布频率是?
A) 单次发布(一次性完成) B) 分阶段发布(多个版本逐步上线) C) 持续发布(多次迭代,持续优化)
步骤 4: 设计回滚方案
使用 AskUserQuestion:
🔙 回滚触发条件
什么情况下需要回滚?
A) 严重Bug导致功能不可用 B) 性能严重下降(响应时间、错误率) C) 用户投诉激增 D) 数据异常(关键指标暴跌) E) 以上全部情况
继续询问:
⏱️ 回滚时间要求
从决定回滚到完成回滚,最长可接受时间:
A) 5分钟内(快速回滚) B) 15分钟内(标准回滚) C) 30分钟内(慢速回滚) D) 1小时内(可接受)
步骤 5: 规划通知机制
使用 AskUserQuestion:
📢 上线通知对象
需要通知哪些人?(可多选)
A) 内部团队(产品、研发、测试、运营) B) 管理层(项目发起人、部门负责人) C) 外部用户(发布公告、更新日志) D) 合作伙伴(第三方服务、渠道方) E) 客服团队(提前准备FAQ)
步骤 6: 生成上线执行方案
使用 Write 工具生成 docs/04-风控管理/上线执行方案.md。
Subagent 并行加速(v2.0.0 新增)
利用 Agent 工具并行执行独立子任务,大幅缩短总执行时间。
可并行子任务
当步骤1-3的用户信息收集完成后,以下两个任务可以并行执行:
| 子任务 | 说明 |
|---|---|
| 检查清单编排 | 基于上线策略和检查项,自动生成完整上线检查清单 |
| 回滚方案设计 | 根据回滚触发条件和时间要求,输出回滚操作步骤 |
触发方式
在步骤6生成文档前,使用 Agent 工具激活子任务并行执行。
V1 vs V2 对比
| 维度 | V1.1.0(串行) | V2.0.0(Subagent并行) | 节省 |
|---|---|---|---|
| 检查清单 | 用户逐一确认检查项 | Agent并行生成完整清单 | 约3轮交互 |
| 回滚方案 | 依次询问回滚细节 | Agent自动输出回滚步骤 | 约2轮交互 |
| 通知机制设计 | 逐个问询通知对象 | Agent并行编排通知方案 | 约2轮交互 |
| 总交互轮次 | 约12-15轮 | 约6-8轮 | 减少50%+ |
| 耗时估算 | 12-18分钟 | 6-9分钟 | 节省约8分钟 |
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.
- 1mo ago Changed · +3 lines e6a72262a47c
- 1mo ago First seen · 670 lines · 60 tokens per session scan A 0da178030423
pm-release is a skill published in the GitHub repository konglong87/superPM (83 stars, last pushed 2d ago), licensed MIT. It adds 60 tokens to every session and 5,061 once invoked, about $0.0002 per session on Opus 5.5. 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
release-skills
A release workflow for preparing and publishing a new version of a software project. A release is a named version that may include updated version files, release notes, tags, or a GitHub Release.
production-build-verification
Build the standalone Forge artifact with strict routing inputs and confirm its static output exists.
trusted-publish
Publish a package from a git repo to its registry via OIDC Trusted Publishing — no long-lived tokens. Writes the GitHub Actions publish workflow, walks the user through the registry-side trust entry, releases, and verifies. Use when the user wants to publish or release a package to PyPI, npm, crates.io, or another…
publish-release
Release moi-computer to npm — verify locally, hand off to the gated GitHub Actions workflow, then verify the published package. Defaults to a next preview; pass stable for a latest release. Use when the user asks to publish, release, ship a version, or cut a dev preview.
multi-search
A web-search helper that chooses among DuckDuckGo, Tavily, Bing API, and Bing scraping based on the available network.
update-nanoclaw
Transactionally update a customized NanoClaw checkout from official upstream without exposing live mounted source, with fork-safe skill refresh, mutable-state snapshots, migration gates, exact-code upgrade markers, detected service restart, health verification, and automatic local rollback. Use for routine merge…