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 agentscope-ai/skills --skill guidance-zhgit clone --depth 1 https://github.com/agentscope-ai/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/agentscope-ai/skills/guidance-zh)<a href="https://agentmods.dev/skills/agentscope-ai/skills/guidance-zh"><img src="https://agentmods.dev/badge/skills/agentscope-ai/skills/guidance-zh/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/agentscope-ai/skills/guidance-zh"><img src="https://agentmods.dev/badge/skills/agentscope-ai/skills/guidance-zh.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 | $0.00044 | $0.01266 |
| Opus 5 | $0.00022 | $0.00633 |
| Sonnet 5 | $0.00009 | $0.00253 |
| Haiku 4.5 | $0.00004 | $0.00127 |
Grade A, and why
guidance scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
This is a copy
95% identical to guidance — 34 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 138 lines — stays where its author put it; the contents beside it link to each section on GitHub.
QwenPaw 安装与配置问答指南
当用户询问 QwenPaw 的安装、初始化、环境配置、依赖要求、常见配置项 时,使用本 skill。
核心原则:
- 先查本地文档,再回答
- 回答要基于已读到的内容,不臆测
- 回答语言与用户提问语言保持一致
标准流程
第一步:定位文档位置
查找记忆中的文档目录
首先你可以查看memory中是否有文档目录,如果有则直接使用,如果没有则继续执行下一步。
# 获取memory中的文档目录
DOC_DIR=$(find ~/.qwenpaw/memory/ -type d -name "docs")
如果 memory 中没有文档目录,则继续执行下面的逻辑。
检查项目源码中的文档目录
执行以下脚本逻辑来获取变量 $QWENPAW_ROOT:
# 获取二进制绝对路径
COP_PATH=$(which qwenpaw 2>/dev/null || whereis qwenpaw | awk '{print $2}')
# 逻辑推导:如果路径包含 .qwenpaw/bin/qwenpaw,则根目录在其上三层
# 例如:/path/to/QwenPaw/.qwenpaw/bin/qwenpaw -> /path/to/QwenPaw
if [[ "$COP_PATH" == *".qwenpaw/bin/qwenpaw" ]]; then
QWENPAW_ROOT=$(echo "$COP_PATH" | sed 's/\/\.qwenpaw\/bin\/qwenpaw//')
else
# 兜底:尝试获取所在目录的父目录
QWENPAW_ROOT=$(dirname $(dirname "$COP_PATH") 2>/dev/null || echo ".")
fi
echo "Detected QwenPaw Root: $QWENPAW_ROOT"
验证并列出文档目录: 使用推导出的 $QWENPAW_ROOT 定位文档:
# 组合标准文档路径
DOC_DIR="$QWENPAW_ROOT/website/public/docs/"
# 检查路径是否存在并列出文件
if [ -d "$DOC_DIR" ]; then
find "$DOC_DIR" -type f -name "*.md" | head -n 100
else
# 如果推导路径不对,执行全局模糊搜索
find "$QWENPAW_ROOT" -type d -name "docs" | grep "website/public/docs"
fi
如果项目文档不存在,搜索工作目录
如果还是找不到文档,搜索 qwenpaw 安装路径下的可用文档内容:
# 寻找 faq.en.md 或 config.zh.md 等特征文件
FILE_PATH=$(find . -type f -name "faq.en.md" -o -name "config.zh.md" | head -n 1)
if [ -n "$FILE_PATH" ]; then
# 使用 dirname 获取该文件所在的目录
DOC_DIR=$(dirname "$FILE_PATH")
fi
如果找到了文档目录,请你记录在 memory 中,格式为:
# 文档目录
$DOC_DIR = <doc_path>
第二步:文档检索与匹配
文档文件命名格式为 <topic>.<lang>.md(如 config.zh.md、config.en.md、quickstart.zh.md)。
使用 find 命令在目标目录中列出所有符合后缀的文档,并根据文件名关键字(如 install, env, setup)锁定目标作为 <doc_path>。
# 列出所有符合后缀的文档
find $DOC_DIR -type f -name "*.md"
如果没有合适的文档,则在下一步阅读所有文档内容。
第三步:阅读文档内容
找到候选文档后,读取并确认与问题相关的段落。可使用:
cat <doc_path>file_readerskill(推荐用于更长文档或分段读取)
如果文档很长,优先读取和问题最相关的章节(安装步骤、配置项、示例命令、注意事项、版本要求)。
第四步:提取信息并作答
从文档中提取关键信息,组织成可执行答案:
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.
- yesterday First seen · 138 lines · 44 tokens per session scan A 41cde0ab8b48
guidance is a skill published in the GitHub repository agentscope-ai/skills (123 stars, last pushed yesterday), licensed Apache-2.0. It adds 44 tokens to every session and 1,266 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. It is 95% identical to guidance, differing in 34 lines, and is treated as a copy.
Other skills, from other repositories
guidance
A guide for answering questions about installing and configuring QwenPaw, a software tool with local documentation. It says to check those documents first and use official website information only when needed.
edu-math-tutorial
A Chinese-language guide for turning a maths problem into a step-by-step teaching video. It explains how to break down the solution, write narration, format equations, and structure scenes.
explainer
A tool for making narrated explainer or tutorial videos with computer-generated visuals. It can help create the script, the video, or both.
code-explanation
Get layered, context-aware explanations of unfamiliar code. Understand what it does, why it was written that way, and how to work with it safely.
teach
Teach the user a new skill or concept, within this workspace.
change-quiz
A self-contained HTML report that explains a set of code changes and ends with a quiz before the changes are merged.