dsh-agent-kit
AGENTS.md0 starsChanged 3 months agoMITView on GitHub
本文件是所有 AI agent(DSH、Claude Code、Cursor 等)操作此仓库时的行为准则。 每次 agent 操作前请先读取本文件。
- Sends data out
Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.
Of the AGENTS.md files we read, the 31,774 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
In this repository12 other files it gives its agents.
- academic-figure-skillSkill
- browserskillSkill
- dsh-fixSkill
- github-repo-verificationSkill
- humanities-thesisSkill
- mcp-builderSkill
- prompt-craftSkill
- scipilot-figure-skillSkill
- study-coachSkill
- systematic-debuggingSkill
- verification-before-completionSkill
- web-researchSkill
The file
We split this AGENTS.md into the rules it gives an agent
1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 58
commit 格式:Conventional Commits(feat: / fix: / docs: / chore:)
≈ “Common” is an estimate: at least 1 in 5 of the 40,246 AGENTS.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
AGENTS.md — dsh-agent-kit 项目规范
本文件是所有 AI agent(DSH、Claude Code、Cursor 等)操作此仓库时的行为准则。 每次 agent 操作前请先读取本文件。
项目概述
- 项目名:dsh-agent-kit — 自用 Agent 工具集
- 技术栈:YAML(MCP 配置)/ Markdown(Skills / 文档)/ PowerShell(安装脚本)
- 用途:把 DSH Desktop 上验证过的插件/MCP/Skills/预设封装成可移植工具集
- 仓库地址:GitHub 上的公开仓库
构建与验证
- 语法检查:
scripts/verify.ps1(校验 YAML 语法、MCP 配置格式、Skill 目录完整性) - 无构建步骤(纯配置/脚本仓库,无代码编译)
- 提交前必须运行
verify.ps1通过
文件规范
| 文件类型 | 规则 |
mcp/*.yml | 符合 dsh-mcp-client 的 cordis.patch.yml insert 格式 |
skills/*/SKILL.md | 必须含 frontmatter(name/description) |
presets/*.yml | 符合 cordis 预设格式 |
scripts/*.ps1 | PowerShell 5.1+ UTF-8 BOM,cmd /c 绕开执行策略 |
docs/*.md | 中文正文,代码/命令保留英文 |
禁止事项(硬约束)
- 禁止提交任何 API key、Token、密码(包括测试值、占位符中的真实值)
- 禁止引入未验证的第三方包——先通过 npm registry 或 GitHub API 验证包名/仓库真实存在
- 禁止修改 DSH 运行配置(
settings.yaml/cordis.patch.yml)除非用户明确要求 - 禁止 push 到 main 前不拉取 —— 先
git pull --rebase origin main - 禁止编造 BV 号 / npm 包名 / 仓库名——所有外部引用必须实测核实
已知坑点(实测验证,勿重蹈)
api.b.ai 网关路径(重要!)
- api.b.ai 只允许推理路径:
/v1/chat/completions、/v1/messages、/v1/responses、/v1/models、/v1/images/* - 请求裸
/v1会返回 HTTP 403:{"message":"HTTP node only allows access to inference API paths (...)", "success":false} - 配置 provider 时必须写完整路径:
- ✅
base_url = "https://api.b.ai/v1"+request_url = "https://api.b.ai/v1/chat/completions" - ❌
request_url = "https://api.b.ai/v1"(裸路径,403)
- ✅
- 验证命令:
curl -X POST https://api.b.ai/v1/chat/completions -H "Authorization: Bearer $KEY" -d '{"model":"deepseek-v4-flash","messages":[{"role":"user","content":"hi"}]}'
PowerShell 5.1 中文脚本
.ps1含中文必须 UTF-8 BOM,否则解析为 GBK 报"字符串缺少终止符"- 正则
{20,}在 PS 5.1 字符串中解析崩溃 → 用{20}或避免花括号量词
MCP 首次启动
- npx 首次运行下载包较慢(120s 超时属正常)
- anki-mcp-server 仅在 Anki 运行时注册工具;bilibili-mcp 需先
config配 Cookie
Git 规范
- commit 格式:Conventional Commits(
feat:/fix:/docs:/chore:) - 分支:个人项目直接 main,谨慎提交
- 发布:打 tag
v1.0.0并创建 GitHub Release
安全规则
- 安装脚本
install.ps1必须先备份目标文件(*.bak-<timestamp>) - 所有对外网络请求必须通过 HTTPS,不发送本地密钥
- CI 工作流使用 GitHub Secrets,不硬编码密钥
触发条件
- 用户说"安装"/"部署"/"恢复环境" → 运行
scripts/install.ps1 - 用户说"校验"/"检查" → 运行
scripts/verify.ps1 - 用户说"导出当前配置" → 运行
scripts/export.ps1 - 用户说"发布"/"release" → 打 tag + GitHub Release
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

