Decides when one particular conversation should kick in, by editing the example phrases people might type and the plain description the AI reads when it chooses.
WHEN TO USE
SCOPE
USE FOR
Reword the description the AI reads when it decides which conversation to open.
Add more of the ways people actually word a request, so fewer of them get missed.
Throw out a weak set of phrases and put a better-spread one in its place.
NOT FOR
Cannot change a topic's trigger type — that requires recreating the topic.
Cannot add trigger phrases to system topics that don't use recognized-intent triggers.
This is a saved copy of the skill's SKILL.md instructions. It may differ from the upstream file or the version you install.
Edit Topic Triggers
Modify how a topic gets triggered: trigger phrases (triggerQueries) and model description (modelDescription).
Instructions
Auto-discover the agent directory:
Glob: **/agent.mcs.yml
NEVER hardcode an agent name.
Find the target topic:
Glob: <agent-dir>/topics/*.topic.mcs.yml
If $ARGUMENTS specifies a topic name, match it. Otherwise, list all topics and ask which one.
Read the topic file and identify what to change.
Check the agent's settings.mcs.yml for GenerativeActionsEnabled. This determines which trigger mechanism matters most.
Make the requested changes using the Edit tool.
Model Description (modelDescription)
The modelDescription is a natural language description at the root of an AdaptiveDialog that tells the generative orchestrator when to route to this topic. When GenerativeActionsEnabled: true, this is the primary way the agent decides which topic to trigger — even more important than trigger phrases.
kind: AdaptiveDialog
modelDescription: This topic helps users order a pizza by collecting their preferred size, toppings, and delivery address.
beginDialog:
kind: OnRecognizedIntent
id: main
...
Best Practices for Model Description
Write a clear, specific sentence describing what the topic does and when it should be used
Focus on the user's intent: "Use this topic when the user asks for information about our perfumes" rather than "This topic sends a message"
Include key scenarios the topic covers: "This topic handles billing inquiries, refund requests, and payment issues"
Keep it concise but complete — the orchestrator reads this to decide routing
Avoid vague descriptions like "This topic handles requests" — be specific about which requests
Examples from Real Topics
# Good — specific about what the topic does and when to use it
modelDescription: Use this topic when the user asks for help or needs assistance with something. Search the SharePoint knowledge base to find relevant information.
# Good — describes the full flow
modelDescription: This topic takes the user's favorite color, generates a word by concatenating the color with four random digits, and returns the generated word. If the color is white, it also provides information about the wavelength.
# Good — describes the computation
modelDescription: This sample topic performs the sum between the two numbers in input
When to Add/Edit Model Description
Always add a modelDescription to topics with OnRecognizedIntent when the agent has GenerativeActionsEnabled: true
Update it when the topic's purpose or scope changes
Topics with system triggers (OnError, OnUnknownIntent, OnConversationStart) don't need a modelDescription — they're triggered automatically
Trigger Phrases (triggerQueries)
Trigger phrases are used for intent recognition. They live inside beginDialog.intent.triggerQueries.
Only topics with OnRecognizedIntent triggers have editable trigger phrases. Other trigger types (OnConversationStart, OnUnknownIntent, etc.) are system-triggered and don't use phrases.
Operations
Add phrases: Append new entries to triggerQueries
Remove phrases: Delete entries from triggerQueries
Replace phrases: Swap old entries for new ones
Rewrite all: Replace the entire triggerQueries array
Best Practices for Trigger Phrases
Use 5-10 phrases per topic for good recognition
Cover synonyms and variations (e.g., "help", "assist", "support", "I need help")
Include different sentence structures (questions, statements, keywords)
Avoid overlap — don't use the same phrases in multiple topics
Keep phrases short and natural (how users actually type/speak)
Mix formal and informal phrasing
Example Structure
kind: AdaptiveDialog
modelDescription: This topic helps users with greeting and introduction.
beginDialog:
kind: OnRecognizedIntent
id: main
intent:
displayName: Greeting
triggerQueries:
- Hello
- Hi
- Hey
- Good morning
- Good afternoon
What This Skill Cannot Do
Cannot change the trigger type (e.g., from OnRecognizedIntent to OnConversationStart) — this requires recreating the topic.
Cannot add triggers to system topics that don't use OnRecognizedIntent.
No requirements beyond installing this skill were found.
🔌 This skill ships inside the Copilot Studio YAML Toolkit plugin. Listed package contents include 4 agents and hooks. Installed components and skill versions may vary by AI tool.
Install just this skill
Run this outside your AI tool, using Node.js 22.20.0 or newer. Skills CLI 1.7.2 skips skill selection; prompts cancel installation without an interactive terminal. If choosing Project when asked about installation scope, open your project's folder before running this.
Install checklist
Check selected tools if a picker appears; Claude Code may be unchecked under Additional agents. ↑/↓ moves, Space toggles selections, Enter confirms.
Without a tool picker, check Installing to: for your intended tool. Stop if it names a different tool.
If asked, choose Symlink (Recommended) for shared files or Copy to all agents for separate copies; review the summary before confirming installation.
Install the find-skills skill? defaults to Yes, adding another skill for all projects. Choose No if you don't want that extra installation.
Skills CLI 1.7.2 sends install data to skills.sh by default, including the source, skill names, target tools and file details; GitHub sources are reported only when GitHub confirms the repository is public, while other remote sources can also be reported. Confirmed-public GitHub sources also trigger a security-audit request. Before installing, set the environment variable DO_NOT_TRACK=1 or DISABLE_TELEMETRY=1 to disable both.
After installing for Claude Code, open it in the installation project if you chose Project; Global makes the skill available across projects on this computer. Once Claude Code loads the skill, its description helps Claude decide when to use it.
If installation created a top-level skills folder that did not exist when your Claude Code session started, run /reload-skills in that session. Run it again after later changes in that folder.
Install the whole plugin
This route installs the Copilot Studio YAML Toolkit plugin package selected for your AI tool.
Add Microsoft's marketplace, then install the plugin
Type both lines inside a Claude Code session, not at your shell prompt. They add Microsoft's marketplace, which is this repository, and the copilot-studio plugin, with its Copilot Studio skills and four sub-agents.
Install checklist
Microsoft lists Node.js 18 or later, and VS Code with the Copilot Studio extension for cloning, pushing and pulling agents.
Claude Code installs it once you pick an install option.
Good to know
A hook that runs whenever a session starts adds Microsoft's Copilot Studio instructions; one tells Claude to lean towards reading your requests as Copilot Studio requests unless there are clear signs otherwise.
The option that installs it for you in every project brings the hook's instructions into every project; to keep them to one Copilot Studio repository, pick the option for this repo only.
On first run, and after an update that changes its dependencies, the session-start hook runs npm install in the plugin's data folder; every session it also writes ~/.copilot-studio-cli/plugin-paths.json.
Add Microsoft's marketplace, then install the plugin
Type both lines in an interactive GitHub Copilot CLI session, not at your shell prompt. The first adds Microsoft's marketplace, which is this repository; the second installs the copilot-studio plugin.
Install checklist
Microsoft lists Node.js 18 or later, and VS Code with the Copilot Studio extension for cloning, pushing and pulling agents.
The plugin also carries a hooks/hooks.json file with SessionStart hooks, a path GitHub's plugin reference lists as a hooks file for plugins in this format; read the file in the repository before you install.
Guide users through adding a new connector action to a Copilot Studio agent. Connector actions require UI-based connection setup, so this skill walks users through the Copilot Studio portal steps, then delegates to edit-action for YAML modifications.Read full description
Generate and insert an Adaptive Card into a Copilot Studio topic using AdaptiveCardPrompt. Use when the user asks to add an adaptive card, rich card, form card, info card, confirmation card, or interactive card to a topic.Read full description
Add generative answer nodes (SearchAndSummarizeContent or AnswerQuestionWithAI) to a Copilot Studio topic. Use this instead of /add-node when the user asks to add grounded answers, knowledge search, generative answers, or AI-powered responses — these nodes require specific patterns (ConditionGroup follow-up, knowledge source references, autoSend, responseCaptureType) that /add-node does not cover.Read full description
Add a global variable to a Copilot Studio agent. Use when the user needs a variable that persists across topics in the same conversation and can optionally be visible to the AI orchestrator.Read full description
Add a knowledge source (public website or SharePoint) to a Copilot Studio agent. Use when the user asks to add a knowledge source, documentation URL, website, or SharePoint site for the agent to search.Read full description
Add or modify a node in an existing Copilot Studio topic. Use when the user asks to add a question, message, condition, variable, or other node to a topic. Do NOT use this for generative answers or knowledge search — use /add-generative-answers instead.Read full description
