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 yyyyolo7a79-sketch/claude-conventions-skill --skill conventionsgit clone --depth 1 https://github.com/yyyyolo7a79-sketch/claude-conventions-skillWrote 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/yyyyolo7a79-sketch/claude-conventions-skill/conventions)<a href="https://agentmods.dev/skills/yyyyolo7a79-sketch/claude-conventions-skill/conventions"><img src="https://agentmods.dev/badge/skills/yyyyolo7a79-sketch/claude-conventions-skill/conventions/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/yyyyolo7a79-sketch/claude-conventions-skill/conventions"><img src="https://agentmods.dev/badge/skills/yyyyolo7a79-sketch/claude-conventions-skill/conventions.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 | $0.01652 |
| Opus 5.5 | $0.00041 | $0.00661 |
| Sonnet 5.5 | $0.00020 | $0.00330 |
| Haiku 4.5 | $0.00010 | $0.00165 |
Grade A, and why
conventions 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 2mo 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 — 146 lines — stays where its author put it; the contents beside it link to each section on GitHub.
全局行为规范
以下三条规则在所有对话中默认生效,除非用户明确要求例外。
规则 1:代码注释默认中文
生成任何代码(Python、Java、JS、SQL、Shell 等)时,注释、docstring、文档字符串默认使用中文。
# ✅ 正确 — 中文注释
def calculate_score(answers: list[dict]) -> float:
"""
根据答题结果计算最终得分。
每道题权重相同,答对 +1,答错 +0,未答跳过。
"""
score = 0.0
for item in answers:
if item.get("correct"): # 仅统计已答且正确的题目
score += 1.0
return score
# ❌ 错误 — 未经要求就写英文注释
def calculate_score(answers):
# Calculate final score based on answers
score = 0.0
...
例外:仅当用户明确说"注释写英文""用英文注释""comments in English"时,才使用英文注释。
规则 2:Bash 命令必须先解释再请求执行
每次向用户展示终端命令时,必须附带中文解释,说明:
- 这条命令做什么(功能描述)
- 为什么需要执行它(意图)
格式:
**这条命令的作用**:<一句话中文说明>
**为什么需要**:<一句话中文说明>
```bash
<命令>
```
实际示例:
**这条命令的作用**:安装 psycopg2-binary Python 数据库驱动包
**为什么需要**:本地脚本需要通过此驱动连接 PostgreSQL 数据库读写数据
```bash
python3 -m pip install psycopg2-binary
```
当需要用户审批权限时,解释尤为重要——让用户不需要读 bash 代码就能判断该不该批准。
适用范围:需要用户审批的命令、有副作用的命令(写入/删除/安装/网络请求)。
连续的只读探查命令(ls/cat/grep/语法检查)只需在一组的开头说明一次意图,
不必逐条重复——否则批量排查时解释文字会淹没实际结果。
规则 3:英文步骤附中文翻译
当你输出英文技术步骤、名词、流程时,在括号中附带中文翻译/注释,方便用户快速理解。
# ✅ 正确
Step 1: Extract the error-fix chain from the conversation.(从对话中提取错误-修复链)
Step 2: Classify each error by category — env / type / logic / external.(按类别分类)
Step 3: Write confirmed items to PgSQL and Obsidian.(写入数据库和笔记)
# ❌ 错误
Step 1: Extract the error-fix chain from the conversation.
Step 2: Classify each error by category — env / type / logic / external.
Step 3: Write confirmed items to PgSQL and Obsidian.
这适用于:
- 技术步骤/流程说明
- 工具/库名称的首次提及(如 "PostgreSQL(关系型数据库)")
- 架构图中的节点和箭头标注
例外:纯代码块(``` 包裹的内容)不需要翻译代码本身;公认的技术缩写(API、JSON、SQL、HTTP)不需要翻译。
规则 4:思考过程使用中文
你的内部推理(thinking / reasoning)使用简体中文。
ALL thinking/reasoning MUST be in Simplified Chinese. Code snippets, error messages, and file paths within thinking can remain in their original language — but your own reasoning text must be Chinese.
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.
- 2mo ago First seen · 146 lines · 102 tokens per session scan A 8583226cc641
conventions is a skill published in the GitHub repository yyyyolo7a79-sketch/claude-conventions-skill (4 stars, last pushed 3mo ago), licensed MIT. It adds 102 tokens to every session and 1,652 once invoked, about $0.0004 per session on Opus 5.5. 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 2026-08-31.
conventions compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| conventionsyyyyolo7a79-sketch/claude-conventions-skill | Skill | 4 | 3mo ago | MIT | A | 102 | Claude Code, Codex |
| odaiorziz/odai | Skill | 133 | 4d ago | MIT | A | 52 | Codex |
| renweiorziz/odai | Skill | 133 | 4d ago | MIT | A | 109 | Codex |
| ribaoorziz/odai | Skill | 133 | 4d ago | MIT | A | 64 | Codex |
| odai-orchestrationorziz/odai | Skill | 133 | 4d ago | MIT | A | 60 | Codex |
| opportunity-skillQuestMeet/opportunity-skill | Skill | 23 | 9d ago | MIT | A | 177 | Claude Code, Codex |
| dsh-run-outputcpj-dev/dsh-plugin-cc | Skill | 104 | 2mo ago | MIT | A | 40 | Claude Code |