Dayward AIWeChat Mini Program
Agent Skills in 7 Days: Turn Experience Into Reusable Capability
Week 1 · D6🔒 LockedAbout 5 hours

Organizing and Distributing: Plugins and Marketplaces, Versioning and Team Sharing, and the Division of Labor Between Function Calling, MCP, and Skills

Go from one person's folder to a team's capability pack: how to organize a plugin, how to publish to a marketplace, how to version it, and exactly what function calling, MCP, and Skills each handle and when to reach for which.

Today's goals 0/3

Sign in to tick these off and save your progress.

Today's Goals

  1. Package a group of skills into a plugin with a manifest and a version, and load it locally to verify it
  2. Explain the three paths for sharing skills across a team and when each fits
  3. Use a comparison table to explain the division of labor between function calling, MCP, and Skills, and give selection criteria

Everything you built over five days still lives only on your own machine. Today you hand it over: how a set of skills becomes one package, how it is published, how versions get bumped, and how a dozen people share one copy. Then we settle the debt this course has owed you since day one — what function calling, MCP, and Skills each actually govern. Once you have read the walkthrough and finished the lab, scroll back to the top and tick off the three goals.

Plain-Language Walkthrough

From one folder to a package

Back to the filing cabinet. You now hold five or six sets of work instructions, and leaving them loose is workable. But the moment you hand them to the department next door, problems appear: they have a folder called "quarterly report" too, so the names collide; you later changed two of the pages and they have no idea whether to update; they only want three of the folders and the other two are irrelevant to them.

Those three problems — namespacing, versioning, boundaries — are everything packaging solves. A package is not a zip of some folders; it is a declaration that these ones are a set, handed over together and upgraded together.

When should things merge into one package? The criterion is whether they are adopted together and retired together. Three of them orbiting the same set of team conventions, where anyone installing one must install all, are one package. One team convention plus one personal debugging habit thrown together only forces somebody to accept the half they did not want.

Conversely, when should one skill split into two? The criterion came on day three, and here is only its package-level corollary: if two parts' trigger situations do not overlap, they should not share one description. Packaging cannot rescue a skill with a blurry boundary; it only amplifies the blur across the whole team.

Sign in and link WeChat to read

Sign in, then link your WeChat mini program account on the account page. Linked accounts share one free article per day. Unlock additional articles with rewarded videos in the mini program.

Mini program QR code

Scan in WeChat to open Dayward AI. Open the same course and chapter, choose the rewarded video, then refresh this page after completion.

Interview questions

  • How do function calling, MCP and Agent Skills relate, and when do you use which?函数调用、MCP 和 Skills 三者的关系是什么?什么时候用哪个?
    Common in ChinaCommon overseasIntermediate#agent-skills#mcp#tool-calling#architecture

    🔒 Unlock this chapter to see the answer keys.

  • A team needs to share more than a dozen skills. How would you organize and distribute them?一个团队要共享十几个 skill,你会怎么组织和分发?
    Common in ChinaCommon overseasIntermediate#agent-skills#distribution#team-governance

    🔒 Unlock this chapter to see the answer keys.

  • Should a skill package be versioned, and what goes wrong most often on upgrade?skill 包要不要做版本管理?升级时最容易出什么问题?
    Common in ChinaCommon overseasDeep dive#agent-skills#versioning#distribution

    🔒 Unlock this chapter to see the answer keys.

Comments