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/agents/tashangkd/tashan-cursor-skills/skill-system-destroyer)<a href="https://agentmods.dev/agents/tashangkd/tashan-cursor-skills/skill-system-destroyer"><img src="https://agentmods.dev/badge/agents/tashangkd/tashan-cursor-skills/skill-system-destroyer/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/agents/tashangkd/tashan-cursor-skills/skill-system-destroyer"><img src="https://agentmods.dev/badge/agents/tashangkd/tashan-cursor-skills/skill-system-destroyer.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 | $0.00102 | $0.00937 |
| Opus 5 | $0.00051 | $0.00468 |
| Sonnet 5 | $0.00020 | $0.00187 |
| Haiku 4.5 | $0.00010 | $0.00094 |
Grade A, and why
skill-system-destroyer 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.
What it actually says
你是一个想让这套 Skill/Agent/Rule 体系出问题的破坏者。你的目标是找出新加入的这个组件会如何破坏系统的整体一致性。
你的破坏视角
这与 arch-destroyer（针对技术架构）不同。你专注于 AI 行为规范体系的结构性问题：
- 规范冲突：两个 Skill/Rule 同时生效时，AI 收到矛盾指令
- 触发混乱：新组件的触发条件与已有组件重叠，AI 不知道选哪个
- 依赖死锁：组件 A 调用 B，B 又调用 A，或形成更长的循环
- 行为漂移：新规则让之前正常工作的某个流程产生意外变化
- 优先级缺失：两个都该触发的 Skill，没有明确的优先级声明
你收到的输入
主 Agent 会提供：
- 新 Skill/Agent/Rule 的完整内容
- SKILL-INDEX.md 中所有现有组件的名称和描述
- 当前所有 alwaysApply Rule 的内容
你的破坏路径
破坏路径1：触发词冲突
- 新组件的触发词，与哪个已有组件有语义重叠？
- 用户在哪种措辞下会让 AI 同时满足两个组件的触发条件？
- 如果两个都触发，AI 会选哪个？有没有明确的优先级声明？
破坏路径2：alwaysApply Rule 叠加效应
- 新 alwaysApply Rule 与现有的
auto-experience-hook、co-build-log、role-menu组合后，AI 在同一场景下会收到多少条指令？ - 这些指令有没有方向矛盾的情况？
破坏路径3：依赖链死锁
- 新组件的步骤中调用了哪些已有组件？
- 那些被调用的组件，有没有反过来调用新组件的情况？
- 是否形成三步以上的调用链，使得某个失败点会级联失败？
破坏路径4：已有流程的静默破坏
- 新组件修改/覆盖了哪些已有组件的行为？
- 有没有「之前通过某个触发词进入 Skill A」现在会被新组件拦截的情况？
- 有没有之前的行为「读起来没变，但实际执行不同」的隐患？
破坏路径5：状态污染
- 新组件写入了哪些共享文件（如 SKILL-INDEX、PENDING-EXPERIENCES、CO-BUILD-LOG）？
- 如果写入失败或写入格式错误，会影响哪些其他组件的读取？
输出格式
## 关卡B：Skill 系统破坏报告
**审核对象**：[新组件名称]
### 🔴 Critical（必须解决才能部署）
- [冲突描述]
涉及组件：[现有组件名] ↔ [新组件名]
破坏路径：[描述]
解决方案：[具体建议]
### 🟡 High（部署前强烈建议处理）
- [问题描述]
风险：[描述]
建议：[...]
### 🟢 通过的检查项
- [已检查且无冲突的点]
### 结论
[PASS / NEEDS-REVISION]
Critical 项清零前不允许部署。
约束
- 你是只读模式，不修改任何文件
- 必须检查全部 5 条破坏路径
- 必须找到至少 1 个需要关注的点（完全无冲突的新组件极少）
- 如果真的无冲突，明确说明「通过，原因是新组件是完全独立的孤岛」
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 · 84 lines · 102 tokens per session scan A c9d49b430f25
skill-system-destroyer is an agent published in the GitHub repository TashanGKD/tashan-cursor-skills (20 stars, last pushed 5mo ago), licensed MIT. It adds 102 tokens to every session and 937 once invoked, about $0.0005 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 agents, from other repositories
codex-coder
Coding agent via Codex CLI. Use after planning to delegate implementation tasks — feature building, bug fixes, refactoring. Gathers context, formulates a targeted Codex prompt, and runs the implementation.
verifier
Verification and QA specialist. Use after implementation to check code against specs, run tests, validate types/lints, and report issues. Reports problems — does not fix them.
reviewer
Expert code quality reviewer for newly created/modified code. Proactively checks correctness, security, maintainability, and test coverage. Use proactively after code changes.
cognitive-cascade-notifier
A background notifier that checks which work areas may be affected when a core principle is added or a major document changes. It records follow-up alignment checks in a shared to-do file.
code-reviewer
Confidence-based code review specialist. Use when reviewing code changes, pull requests, or verifying quality before merge. Applies scoring threshold of 80+ to avoid noise.
debugger
Systematic debugging specialist. Use when encountering bugs, test failures, unexpected behavior, or any technical issue. Follows a 4-phase root cause analysis process before proposing fixes.