Standardized capabilities
维护者通常 1 天内回复
还没有人认领这个 Issue。
评估
- 难度
- 5/5
- 预计耗时
- 一周以上
- 新手友好度
- 28/100
- Issue 类型
- 功能
- 描述清晰度
- 需要澄清
- 活跃度
- 冷清
- 技术栈
- web-dev
- 领域
- api
调研方向
Start with the issue and the referenced get_weather explainer example, then locate where WebMCP tool definitions are specified in the Bikeshed source. The proposal needs a concrete capability-identifier syntax, compatibility rules, and whether aliases or mappings should be supported. Done means the specification clearly defines how an API declares a standardized capability and how consumers recognize it.
由索引模型根据 Issue 内容生成。
描述
I know that one of the core promises of MCP (and WebMCP) is that the Magic Computer will work out how to decode the interface documentation to drive it. But that's inefficient and there will be common patterns that evolve. Sites request the same information over and over in ways that are 99% the same.
Can there be a way to annotate an API to say that it conforms to a known, standardized interaction pattern? Even if that pattern is a subset of the API (there might be proprietary additions to it), being able to say "I understand interaction pattern as standardized in X" would be of great value. If WebMCP is about efficiency, this would allow for even greater efficiency.
For example, a site might offer a simple interface switchLightDarkMode() that takes values from ["light", "dark", "detect"]. This could be standardized and given an identifier of some sort. The API could declare that it supports that identifier.
Take the "get_weather" example in the explainer. That's a generic capability provided by many sites. If they collectively decide to standardize, this could be spelled out as a new property on the tool definition. A minimal version would just say "this interface conforms to standard X".
A more sophisticated version would allow the support to be declared despite differences in spelling. If the standard uses "getWeather" rather than "get_weather", there could be rules for mapping from the standard to the true interface.
Tools within the browser might then be able to detect the identifier and activate, connecting to the capability that is exposed. No special parsing of the interface is then required if the "standardized" interface is understood. "Standards" could be the old-fashioned kind, where someone writes a specification. Or maybe they could be a reference to another MCP/tool interface, hopefully one that is widely understood.
- 主要语言
- Bikeshed
- 星标
- 4.7k
- 派生
- 328
- 平均合并
- 2 天 15 小时
- 30 天内合并 PR
- 16
环境准备
这个项目没有提供开发容器、Dockerfile 或贡献指南，环境需要你自己搭建：先看它的 README，通用步骤见我们的新手贡献指南。
从这里开始
- 先读完整个 Issue，再读项目的贡献指南。
- 在 Issue 下留言说明你要接手 —— 这能避免两个人做同样的事。
- Fork 仓库，在一个分支上完成修改。
- 提交 Pull Request，并在描述里引用这个 Issue 编号。
webmachinelearning/webmcp 的其他 Issue
-
难度 1/5 1-3 小时 新手友好度 72/100
webmachinelearning/webmcp#360 ·
维护者通常 1 天内回复
-
Say how an agent should weigh annotations that the page wrote about itself可能已有人在做 关联的 PR 仍在进行中或已合并。 未关闭
难度 2/5 1-3 小时 新手友好度 72/100
webmachinelearning/webmcp#353 ·
维护者通常 1 天内回复
-
难度 2/5 1-3 小时 新手友好度 67/100
webmachinelearning/webmcp#343 · 1 个 reaction ·
维护者通常 1 天内回复
-
Agenda+ i18n-tracker
难度 2/5 1-3 小时 新手友好度 75/100
webmachinelearning/webmcp#273 · 3 条评论 ·
维护者通常 1 天内回复
-
a11y-tracker Agenda+
难度 2/5 1-3 小时 新手友好度 67/100
webmachinelearning/webmcp#272 · 4 条评论 · 2 个 reaction ·
维护者通常 1 天内回复
查看 webmachinelearning/webmcp 的全部 Issue
相似的 Issue
-
难度 2/5 1-3 小时 新手友好度 82/100
volcengine/OpenViking#6445 ·
维护者通常 1 天内回复
-
难度 2/5 1-3 小时 新手友好度 65/100
维护者通常 1 天内回复
-
难度 2/5 1-3 小时 新手友好度 72/100
EnzymeAD/Reactant.jl#3549 ·
维护者通常 1 天内回复
-
难度 2/5 1-3 小时 新手友好度 74/100
deepspeedai/DeepSpeed#8884 ·
维护者通常 3 天内回复
-
OCIRepository: pick the Helm chart layer automatically when the first layer is a provenance file未关闭
难度 2/5 1-3 小时 新手友好度 78/100
fluxcd/source-controller#2198 ·
维护者通常 1 天内回复