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.
git clone --depth 1 https://github.com/TashanGKD/tashan-cursor-skillsWrote 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/rules/tashangkd/tashan-cursor-skills/comprehensive-change-guard)<a href="https://agentmods.dev/rules/tashangkd/tashan-cursor-skills/comprehensive-change-guard"><img src="https://agentmods.dev/badge/rules/tashangkd/tashan-cursor-skills/comprehensive-change-guard/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/rules/tashangkd/tashan-cursor-skills/comprehensive-change-guard"><img src="https://agentmods.dev/badge/rules/tashangkd/tashan-cursor-skills/comprehensive-change-guard.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 | $0.01794 | $0.01794 |
| Opus 5 | $0.00897 | $0.00897 |
| Sonnet 5 | $0.00359 | $0.00359 |
| Haiku 4.5 | $0.00179 | $0.00179 |
Grade A, and why
comprehensive-change-guard scanned grade A with 0 findings 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 10d 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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 122 lines — stays where its author put it; the contents beside it link to each section on GitHub.
综合变更协调守门规则(comprehensive-change-guard)
根因:AI 的触发词扫描(A1)在多层变更场景下只命中最显眼的单角色, 导致其他层次被跳过,关卡A/B/C被绕过。 本 Rule 在 A1 之后补充「多层检测」步骤,强制按序执行对应角色。
证据来源:产品开发域 75 个沙盘 Phase 2 验证(PD-016~PD-075)根因一「缺编排层」
- domain-optimizer-产品开发-20260323.md P0-5
触发条件(A1 触发词扫描之后执行)
消息语义同时命中以下 ≥ 2 个层次,且动作类型为「变更实施/方案设计」(而非纯问答/分析):
□ 产品层:「功能」「用户流」「产品方案」「闭环」「需求」「产品设计」
□ 架构层:「技术框架」「接口」「数据模型」「架构」「技术方案」「技术设计」
□ UI层:「界面」「交互」「设计稿」「视觉」「组件设计」
□ 代码层:「实现」「写代码」「前端开发」「后端开发」「AI工程」「代码细节」
排除场景(不触发):
- 纯问答/分析:「这个设计合理吗?」「怎么看这个技术选型?」
- 单层变更:「帮我写用户注册接口」(只有代码层)
- 已在单角色 Skill 执行流程中的中间步骤(防止嵌套触发)
触发后立即执行
Step 1 声明多层变更检测(必须在第一句输出,不可省略):
「⚠️ [多层变更检测] 本任务涉及 [X层 + Y层(+ Z层)],
需按以下顺序执行,不可跳过关卡:」
Step 2 输出本次实际涉及层次的执行序列(只列实际涉及的层,不列不相关的):
有产品层变更 → ① PM-IN(role-产品经理) → 关卡A(role-审核者-用户模拟)
有架构层变更 → ② ARCH(role-技术架构师) → 关卡B(role-审核者-系统破坏)
有UI层变更 → ③ UI(role-UI设计师) [在关卡B通过后执行]
有代码层变更 → ④ FE/BE/AI-ENG 对应角色 → 关卡C(role-测试工程师)
示例输出:
「执行顺序:
① role-产品经理 → 关卡A
② role-技术架构师 → 关卡B
③ role-后端开发 → 关卡C
现在开始第①步,请确认后触发 role-产品经理。」
Step 3 等待用户触发第①步角色(不允许自动跳到后面的层次)
Step 4 每完成一层、走完对应关卡后,提示下一层:
「✅ [层次名]已完成并通过关卡[X]。下一步:[下一层角色]」
禁止行为(硬性约束)
❌ 禁止:一次性输出跨越多个层次的综合方案(无论看起来多「完整」)
❌ 禁止:以「改动很小」为由跳过关卡直接进代码
❌ 禁止:在产品层变更未经关卡A时进行架构设计
❌ 禁止:在架构层变更未经关卡B时开始代码实现
❌ 禁止:将「技术性很强」作为跳过产品层的理由
与已有规则的关系
| 相关规则/Skill | 关系 |
|---|---|
session-bootstrap.mdc A1 |
接续:本 Rule 在 A1 触发词扫描之后执行(A1.10),不替代 A1 |
role-menu.mdc 规则 2-4(关卡序列) |
强化:本 Rule 在「多层场景」中强制执行这些规则,不冲突 |
cognitive-ask |
互补:用户若先说「基于我的文档」,则先触发 cognitive-ask 读文档,再触发本 Rule 做变更编排 |
frontend-brand-guard.mdc |
同级:同为 alwaysApply 守门 Rule,本 Rule 守「变更编排」,对方守「品牌规范」 |
认知根
本 Rule 的认知根:
- L1 文档:
AI时代产品问题全景框架.md(§产品开发流程的完整性原则) - L1.5 原则:P12「最小必要限制」(每个关卡只守自己的门,但都必须守)
- 沙盘证据:PD-016~PD-075(75 个沙盘,根因一:缺编排层)
变更记录
v1.0 — 2026-03-23 — 初始创建
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.
- 10d ago First seen · 122 lines · 1,794 tokens per session scan A 55e9e59f99dd
comprehensive-change-guard is a cursor rule published in the GitHub repository TashanGKD/tashan-cursor-skills (20 stars, last pushed 5mo ago), licensed MIT. It adds 1,794 tokens to every session, about $0.0090 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other cursor rules, from other repositories
10-feature-development
Feature implementation workflow and engineering mindset.
11-template-conventions
Reusable building blocks shipped with this template - use them instead of writing new ones.
03-ui
UI, layout, theming and localization standards.
09-android-best-practices
Android engineering best practices.
12-new-project
Workflow for starting a new app from this template - rebranding, identity, cleanup and first feature.
13-updating-project
Workflow for updating an existing project - dependency and SDK upgrades, migrations, refactors, bug fixes.