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 sanqi-cd/Sanqi-Skills --skill skill-buildergit clone --depth 1 https://github.com/sanqi-cd/Sanqi-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/sanqi-cd/sanqi-skills/skill-builder)<a href="https://agentmods.dev/skills/sanqi-cd/sanqi-skills/skill-builder"><img src="https://agentmods.dev/badge/skills/sanqi-cd/sanqi-skills/skill-builder/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/sanqi-cd/sanqi-skills/skill-builder"><img src="https://agentmods.dev/badge/skills/sanqi-cd/sanqi-skills/skill-builder.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00090 | $0.02229 |
| Opus 5.5 | $0.00036 | $0.00892 |
| Sonnet 5.5 | $0.00018 | $0.00446 |
| Haiku 4.5 | $0.00009 | $0.00223 |
Grade A, and why
skill-builder 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 28d 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 — 157 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Skill 构建器
目标
把一个想法或已有 Skill 直接推进到结构合规、指令清晰、资源克制、脚本可靠、评测可复现的完整能力包。除非关键信息确实无法推断,不要停在建议或模板阶段。
核心原则
- 先读现状:修改已有 Skill 时,先完整读取
SKILL.md、已引用资源、脚本、测试和仓库约定。 - 只问阻塞问题:能从上下文合理推断的内容直接处理;必须提问时每轮不超过 3 个。
- 渐进披露:主文件保留触发、决策、流程和质量门槛,长知识放 references,确定性操作放 scripts。
- 验证比例随风险增长:至少覆盖正常输入、缺失输入和边界输入;共享脚本需要单元测试。
- 描述决定发现:
description同时写清任务、典型触发、边界和交付结果,不塞入完整工作流。 - 自动推进:用户已经授权实施时,各阶段通过后直接进入下一阶段,不重复索要确认。
工作流程
1. 建立任务契约
从用户输入和现有文件中提取:
- 用户要反复完成的具体任务。
- 典型触发话术与相邻但不应触发的请求。
- 必需输入、可选输入和合理默认值。
- 最终交付物、保存位置和完成信号。
- 依赖的工具、网络、凭据、运行时或应用。
- 至少 3 条可判定的质量标准和主要失败模式。
只有触发边界、交付格式或高风险行为仍不明确时才提问。用户给出完整规格时直接进入设计;用户只给一句模糊想法时,优先询问“最终交付什么”“谁会在什么场景触发”“什么算成功”。
产出一份内部任务契约。只有存在会改变实现方向的互斥选择时才展示给用户确认,不把确认仪式当成固定门槛。
2. 设计最小能力包
先区分目标:可移植的 Agent Skills 包只要求根目录有 SKILL.md;下面是需要适配 Codex 展示并满足本仓库发布门禁时采用的 Sanqi 完整结构,不是开放标准的强制目录:
skill-name/
├── SKILL.md
├── agents/
│ └── openai.yaml # 仅在目标客户端需要时
├── scripts/ # 仅在确定性操作能提高可靠性时添加
├── references/ # 仅放按需读取的领域知识或长规范
├── assets/ # 仅放最终输出会直接使用的模板或素材
└── evals/ # 本仓库的发布约定
├── evals.json
└── trigger-evals.json
不要为了目录完整而创建空目录。脚本、引用和资产必须在主文件中说明何时使用;每个引用从 SKILL.md 直接链接,避免多层引用链。
选择适合的工作流模式时读取 references/design-patterns.md。
3. 编写标准元数据
顶层使用 Agent Skills 支持的字段。最小可移植包需要 name 和 description;license、compatibility、metadata 属于可选字段。本仓库另外要求 MIT 和统一的展示元数据,不能把这些仓库约定称为开放标准的强制要求:
---
name: skill-name
description: >
[做什么;用户何时会需要;典型话术;相邻边界;最终交付什么。]
license: MIT
compatibility: [运行时、工具、网络或应用要求。]
metadata:
author: "author-name"
version: "1.0.0"
---
要求:
name与目录名完全一致,只含小写字母、数字和单连字符,长度不超过 64。description非空且不超过 1024 字符,使用具体名词和用户话术,不依赖只有作者懂的简称。- 自定义展示信息放进
metadata,所有值使用字符串。 - 仅在 Skill 确实需要时声明
allowed-tools;不要用它掩盖未说明的依赖。
目标是 Codex 等使用该文件的客户端,或发布到本仓库时,补充 agents/openai.yaml。本仓库要求其中字符串加引号,default_prompt 明确提到 $skill-name;其他客户端可能忽略此文件。
4. 编写可执行指令
正文使用祈使语气,包含:
- 一句话目标。
- 开始条件和输入缺失策略。
- 有顺序、分支和产出物的工作流。
- 错误、降级、重试和停止条件。
- 可用“通过/不通过”判断的质量标准。
- 最终交付内容、文件路径和已知局限。
What ships with it
4 files 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.
- 28d ago Changed 4309b595c8e3
- 1mo ago First seen · 157 lines · 90 tokens per session scan A 4174ccdfb9c3
skill-builder is a skill published in the GitHub repository sanqi-cd/Sanqi-Skills (28 stars, last pushed 26d ago), licensed MIT. It adds 90 tokens to every session and 2,229 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 2026-08-30.
Other skills, from other repositories
tmux
Use when a CLI task needs an interactive terminal or control of a tmux session.
add-matrix
Add Matrix channel integration via Chat SDK. Works with any Matrix homeserver.
add-resend
Add Resend (email) channel integration via Chat SDK.
add-macos-statusbar
Add a macOS menu bar status indicator for NanoClaw. Shows a bolt icon with a green/red dot indicating whether NanoClaw is running, with Start, Stop, and Restart controls. macOS only.
sandbase
Access 2,000+ AI models and API tools through one MCP interface for inference, media generation, search, scraping, embeddings, social data, and structured retrieval. Use sandbasediscover before building custom integrations or declaring external data inaccessible; prefer an existing dedicated tool or API key when the…
communication
Use before sending @mentions, deciding whether to reply, TASKCOMPLETED, BLOCKED, QUESTION, direct answers, loop prevention, or suppressing low-information acknowledgements.