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 TashanGKD/tashan-cursor-skills --skill project-convention-resolvergit 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/skills/tashangkd/tashan-cursor-skills/project-convention-resolver)<a href="https://agentmods.dev/skills/tashangkd/tashan-cursor-skills/project-convention-resolver"><img src="https://agentmods.dev/badge/skills/tashangkd/tashan-cursor-skills/project-convention-resolver/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/tashangkd/tashan-cursor-skills/project-convention-resolver"><img src="https://agentmods.dev/badge/skills/tashangkd/tashan-cursor-skills/project-convention-resolver.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 | $0.00151 | $0.02482 |
| Opus 5 | $0.00076 | $0.01241 |
| Sonnet 5 | $0.00030 | $0.00496 |
| Haiku 4.5 | $0.00015 | $0.00248 |
Grade A, and why
project-convention-resolver 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 8d 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 — 196 lines — stays where its author put it; the contents beside it link to each section on GitHub.
项目规范问题自动修复（project-convention-resolver）
定位：项目内规范问题的「接收 → 分类 → 路由执行 → 验证 → 归档」一体化 Skill
类比关系：
- issue-tracker → 记录问题（不执行）
- project-convention-resolver → 记录 + 自动路由执行 + 验证
强绑定 Rule：R2 NO_FABRICATION / R3 READ_FIRST / R7 CLOSED_LOOP_REQUIRED
规范问题分类（五类）
A 类：产品规范偏差
→ 功能不符合产品定义中的闭环设计
→ 动线缺失入口/出口
→ 双层设计缺失（只有人层没有智能体层，或反之）
→ 产品文档描述与实际不符
B 类：技术架构规范偏差
→ 接口设计不符合技术架构.md 规范
→ 模块边界不清晰，违反关注点分离
→ 跨模块耦合过紧，扩展性风险
→ Schema/数据模型与规范不一致
C 类：代码规范/实现偏差
→ 命名规范违反（snake_case/camelCase 混用等）
→ 缺少必要的错误处理或测试
→ 安全漏洞（注入/越权/硬编码密钥）
→ 与产品文档描述不符的实现
D 类：文档规范偏差
→ 文档结构不符合项目规范（缺变更记录/版本号等）
→ 多个文档内容漂移（同一内容在不同文档中描述不一致）
→ README 与实际代码状态不符
E 类：Skill/Rule 规范偏差
→ Skill 描述不准确 / 触发词不对
→ Rule 有冲突 / alwaysApply 叠加矛盾
→ Skill 与实际行为不一致
激活后立即执行
Step 1 记录问题（R2：必须基于用户实际描述）
调用 issue-tracker：
- 如果是产品设计类 → 写入 产品经理/产品问题追踪台.md
- 如果是技术实现类 → 写入 技术架构师/技术问题追踪台.md
- 如果是 Skill/Rule 类 → 写入 .cursor/skills/skill-index/PENDING-EXPERIENCES.md
Step 2 Read 相关文档（R3 READ_FIRST）
根据问题类型读取必要的上下文：
A 类 → Read: 产品经理/产品定义.md
B 类 → Read: 技术架构师/技术架构.md
C 类 → Read: 被报告的代码文件（具体路径）
D 类 → Read: 被报告的文档文件
E 类 → Read: 被报告的 SKILL.md 或 .mdc 文件
Step 3 分类 + 确认修复方向
向用户确认：
「检测到 [X类] 规范问题：[问题描述]
修复方向：[一句话说明]
将调用：[对应角色 Skill / 工具]
[确认，立即修复] [调整方向后修复] [只记录，暂不修复]」
若用户选择「调整方向后修复」：
→ 等待用户说明调整内容
→ 重新展示调整后的修复方向
→ 再次确认后再执行
不允许静默执行修复（规范修复可能有误判，必须用户知情）
Step 4 路由执行（按分类路由到对应角色 Skill）
【A 类 → 按 role-产品经理 激活协议执行】
将「规范问题描述 + 相关产品定义章节内容」作为本次 PM 任务的上下文
按 role-产品经理 的步骤修复产品定义文档或设计逻辑
【B 类 → 按 role-技术架构师 激活协议执行】
将「规范问题描述 + 相关架构文档章节」作为上下文
修复技术架构.md，若涉及产品意图，必须先回 PM 确认
【C 类 → 按 role-后端开发 或 role-前端开发 激活协议执行】
将「规范问题描述 + 具体代码文件路径 + 违反的规范条目」作为上下文
修复代码后自动触发「Bug 修复后强制验证闭环」（调用 /verifier）
【D 类 → 按 doc-consolidator 激活协议执行】
将「规范问题描述 + 涉及的文档文件」作为上下文
整合/修复文档，确保单一真源
【E 类 → 加载 skill-rule-修改规范，强制三问+备份+修改+变更记录】
E 类不允许直接修改，必须先回答三问，再备份，再修改
Step 5 验证修复效果（R7 CLOSED_LOOP_REQUIRED）
C 类（代码修复）：
→ 自动调用 /verifier，传入原问题描述作为验证场景
A/B/D/E 类（文档/规范修复）：
→ 人工确认：「已完成修复，请确认以下变更是否符合预期：[变更摘要]」
→ 等待用户确认
→ 若本次修复来源为 skill-system-health-check 报告 → 修复完成后，告知：「建议重新运行 skill-system-health-check 验证该问题已消解」
Step 6 归档关闭
→ 更新追踪台对应条目状态为「已修复（含修复说明）」
→ 告知用户：「规范问题已修复并验证，追踪台已更新。」
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.
- 8d ago First seen · 196 lines · 151 tokens per session scan A 0ac4c8402191
project-convention-resolver is a skill published in the GitHub repository TashanGKD/tashan-cursor-skills (20 stars, last pushed 5mo ago), licensed MIT. It adds 151 tokens to every session and 2,482 once invoked, about $0.0008 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 2026-09-03.
Other skills, from other repositories
cocoreview
CocoReview — structured code review with six-severity findings vocabulary, progressive disclosure architecture, and universal anti-pattern baseline. Invoked via $review [file] [--complexity] [--security] [--architecture] [--language ].
audit-codemod-safety
Read-only audit of a codemod or bulk mechanical transform for behavior-preservation — compiles/lints is not same-behavior. Use when "did this codemod break anything", "audit this bulk refactor", or before merging a mass find-replace. Diff quality → audit-code-review. SQL → plan-data-integrity.
mattpocock-dev-workflows
Use when you need Matt Pocock's engineering workflow patterns: spec synthesis, deep-module design, domain modeling, two-axis code review, or disciplined bug diagnosis. A meta-skill covering five complementary sub-patterns for real engineering — not vibe coding.
audit
(forwward) Aggressively audits and cleans a codebase — removes slop, dead code, and AI-generated noise; fixes DRY/SOLID violations; restructures toward the simplest correct implementation. One commit, no behavior changes. Use when the user says "audit the code", "clean up tech debt", "remove dead code", "deslop"…
code-review
Paranoid staff engineer code review mode. Finds bugs that pass CI but break in production. Use when user says /review, asks for code review, wants to check for bugs, or before merging important changes.
github-repo-triage
Triage and maintain GitHub repositories — assess open issues/PRs, cherry-pick contributions, manage releases, and publish packages. Use when the user wants to catch up on a repo, review open issues, merge community PRs, cut a release, or publish to package registries.