Int Reference logo

Int Reference

Organization
microsoft
int-reference

Reference tables for Copilot Studio YAML authoring: triggers, actions, variables, entities, Power Fx functions, templates. Preloaded by author and advisor agents.

Overview

Publishermicrosoft
Repositoryskills-for-copilot-studio
Skill nameint-reference
Stars
478
Forks
101
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by microsoft on GitHub. Read the source before you install it.

Installation

Install the Int Reference AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/microsoft/skills-for-copilot-studio/tree/main/skills/int-reference
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/microsoft/skills-for-copilot-studio.git /tmp/skills-for-copilot-studio
mkdir -p .claude/skills
cp -r /tmp/skills-for-copilot-studio/skills/int-reference .claude/skills/int-reference
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Int Reference in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Int Reference on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Int Reference is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Copilot Studio YAML Reference

Core File Types

FilePurpose
agent.mcs.ymlMain agent metadata (kind: GptComponentMetadata)
settings.mcs.ymlAgent settings and configuration
connectionreferences.mcs.ymlConnector references
topics/*.mcs.ymlConversation topics (kind: AdaptiveDialog)
actions/*.mcs.ymlConnector-based actions (kind: TaskDialog)
knowledge/*.mcs.ymlKnowledge sources (kind: KnowledgeSourceConfiguration)
variables/*.mcs.ymlGlobal variables (kind: GlobalVariableComponent)
agents/*.mcs.ymlChild agents (kind: AgentDialog)

Trigger Types

Topics with OnRecognizedIntent have two routing mechanisms — which one matters depends on the orchestration mode:

  • modelDescription — used by generative orchestration (GenerativeActionsEnabled: true). The AI orchestrator reads this to decide routing. Primary mechanism for generative agents.
  • Trigger phrases (triggerQueries) — used by classic orchestration. Pattern-matched against the user's utterance. Secondary hints when generative orchestration is enabled.

System triggers (OnConversationStart, OnUnknownIntent, OnError, etc.) fire automatically and don't use either mechanism.

KindPurpose
OnRecognizedIntentTrigger phrases matched
OnConversationStartConversation begins
OnUnknownIntentNo topic matched (fallback)
OnEscalateUser requests human agent
OnErrorError handling
OnSystemRedirectTriggered by redirect only
OnSelectIntentMultiple topics matched (disambiguation)
OnSignInAuthentication required
OnToolSelectedChild agent invocation
OnKnowledgeRequestedCustom knowledge source search triggered (YAML-only, no UI)
OnGeneratedResponseIntercept AI-generated response before sending
OnOutgoingMessageNon-functional (2026-03-15) — exists in schema but does not fire at runtime. Do not use.

YAML-Only Features

These features work at runtime but are not visible in the Copilot Studio UI. Warn users that UI edits may silently remove them.

FeatureNotes
triggerCondition on knowledge sourcesThe UI only exposes this as an on/off toggle (=false to exclude from UniversalSearchTool). Arbitrary Power Fx expressions (e.g., =Global.UserDepartment = "HR") work at runtime but can only be set via YAML. Use with caution. (2026-03-16)

Action Types

KindPurpose
SendActivitySend a message
QuestionAsk user for input
SetVariableSet/compute a variable (Power Fx expression, prefix =)
SetTextVariableSet a text variable using template interpolation ({}). Useful for converting non-text types (e.g., Number) to text: "You have {Topic.Count} items"
ConditionGroupBranching logic
BeginDialogCall another topic
ReplaceDialogReplace current topic
EndDialogEnd current topic
CancelAllDialogsCancel all topics
ClearAllVariablesClear variables
SearchAndSummarizeContentGenerative answers (grounded in knowledge)
AnswerQuestionWithAIAI answer (conversation history + general knowledge only)
EditTableModify a collection
CSATQuestionCustomer satisfaction
LogCustomTelemetryEventLogging
OAuthInputSign-in prompt
SearchKnowledgeSourcesSearch knowledge sources (returns raw results, no AI summary)
CreateSearchQueryAI-generated search query from user input

Connector Actions (TaskDialog)

Connector actions (kind: TaskDialog) invoke external connector operations. They are stored in actions/ and require a connection reference in connectionreferences.mcs.yml.

Use /add-action to create new actions from available connectors. The schema describes the structural properties of TaskDialog and InvokeConnectorTaskAction, but the specific inputs and outputs for each connector operation are connector-specific — use the connector lookup script (connector-lookup.bundle.js) to get the full operation details.

Action Structure

FieldPurpose
kind: TaskDialogIdentifies this as a connector action
inputsInputs: AutomaticTaskInput (AI-provided) or ManualTaskInput (fixed value)
modelDisplayNameDisplay name for AI orchestrator routing
modelDescriptionDescription for AI orchestrator routing
outputsOutput property names returned by the connector
action.kindAlways InvokeConnectorTaskAction for connector actions
action.connectionReferenceLogical name of the connection (registered in connectionreferences.mcs.yml)
action.connectionProperties.modeMaker (maker's credentials) or Invoker (end user's credentials)
action.operationIdThe connector's specific operation identifier
outputModeUsually All — exports all operation outputs

Input Types

Input KindUse WhenNotes
AutomaticTaskInputThe AI orchestrator should provide the value based on contextIncludes description for the AI to understand what to provide
ManualTaskInputA fixed/hardcoded value (e.g., timezone, folder path)Can only hardcode strings. Non-string values (IDs, enums) should be reviewed by the user after pushing

$-Prefixed Property Names (SharePoint, OData)

Some connectors (notably SharePoint) use OData parameters like $filter, $orderby, $top. These require special quoting in TaskDialog YAML — both single and double quotes:

yaml
# TaskDialog (actions/*.mcs.yml) — CORRECT
- kind: ManualTaskInput
  propertyName: "'$filter'"
  value: "Status eq 'Active'"

"'$filter'" means: the outer "" are YAML string delimiters; the inner '' are part of the literal value sent to the runtime. Using $filter, "$filter", or '$filter' alone will fail.

InvokeConnectorAction (inline in topics) uses a different format — the parameters/ prefix with no inner single quotes:

yaml
# InvokeConnectorAction (inside topics) — CORRECT
- kind: InvokeConnectorAction
  operationId: GetItems
  input:
    parameters/$filter: "Status eq 'Active'"

Never mix these two formats.

System Variables

VariableDescription
System.Bot.NameAgent's name
System.Activity.TextUser's current message
System.Conversation.IdConversation identifier
System.Conversation.InTestModeTrue if in test chat
System.FallbackCountNumber of consecutive fallbacks
System.Error.MessageError message
System.Error.CodeError code
System.SignInReasonWhy sign-in was triggered
System.Recognizer.IntentOptionsMatched intents for disambiguation
System.Recognizer.SelectedIntentUser's selected intent
System.SearchQueryAI-rewritten search query (available in OnKnowledgeRequested)
System.KeywordSearchQueryKeyword version of search query (available in OnKnowledgeRequested)
System.SearchResultsTable to populate with custom search results — schema: Content, ContentLocation, Title (available in OnKnowledgeRequested)
System.ContinueResponseSet to false in OnGeneratedResponse to suppress auto-send
System.Response.FormattedTextThe AI-generated response text (available in OnGeneratedResponse)

Variable Scopes

PrefixScopeLifetime
Topic.<name>Topic variableCurrent topic only
Global.<name>Global variableEntire conversation (defined in variables/ folder)
System.<name>System variableBuilt-in, read-only

Global variables are defined as YAML files in variables/<Name>.mcs.yml (kind: GlobalVariableComponent). aIVisibility accepts UseInAIContext (orchestrator can read and reason about the value) or Hidden (orchestrator unaware — use for flags and internal bookkeeping).

Prebuilt Entities

EntityUse Case
BooleanPrebuiltEntityYes/No questions
NumberPrebuiltEntityNumeric inputs
StringPrebuiltEntityFree text
DateTimePrebuiltEntityDate/time
EMailPrebuiltEntityEmail addresses

Power Fx Expression Reference

Only use functions from the supported list below. Copilot Studio supports a subset of Power Fx — using unsupported functions will cause errors.

yaml
# Arithmetic
value: =Text(Topic.number1 + Topic.number2)

# Date formatting
value: =Text(Now(), DateTimeFormat.UTC)

# Conditions
condition: =System.FallbackCount < 3
condition: =Topic.EndConversation = true
condition: =!IsBlank(Topic.Answer)
condition: =System.Conversation.InTestMode = true
condition: =System.SignInReason = SignInReason.SignInRequired
condition: =System.Recognizer.SelectedIntent.TopicId = "NoTopic"

# String interpolation in activity (uses {} without =)
activity: "Error: {System.Error.Message}"
activity: "Error code: {System.Error.Code}, Time (UTC): {Topic.CurrentTime}"

# Record creation
value: "={ DisplayName: Topic.NoneOfTheseDisplayName, TopicId: \"NoTopic\", TriggerId: \"NoTrigger\", Score: 1.0 }"

# Variable initialization (first assignment uses init: prefix)
variable: init:Topic.UserEmail
variable: init:Topic.CurrentTime
# Subsequent assignments omit init:
variable: Topic.UserEmail

Supported Power Fx Functions

These are all the Power Fx functions available in Copilot Studio. Do NOT use any function not on this list.

Math: Abs, Acos, Acot, Asin, Atan, Atan2, Cos, Cot, Degrees, Exp, Int, Ln, Log, Mod, Pi, Power, Radians, Rand, RandBetween, Round, RoundDown, RoundUp, Sin, Sqrt, Sum, Tan, Trunc

Text: Char, Concat, Concatenate, EncodeHTML, EncodeUrl, EndsWith, Find, Left, Len, Lower, Match, MatchAll, Mid, PlainText, Proper, Replace, Right, Search, Split, StartsWith, Substitute, Text, Trim, TrimEnds, UniChar, Upper, Value

Date/Time: Date, DateAdd, DateDiff, DateTime, DateTimeValue, DateValue, Day, EDate, EOMonth, Hour, IsToday, Minute, Month, Now, Second, Time, TimeValue, TimeZoneOffset, Today, Weekday, WeekNum, Year

Logical: And, Coalesce, If, IfError, IsBlank, IsBlankOrError, IsEmpty, IsError, IsMatch, IsNumeric, IsType, Not, Or, Switch

Table: AddColumns, Column, ColumnNames, Count, CountA, CountIf, CountRows, Distinct, DropColumns, Filter, First, FirstN, ForAll, Index, Last, LastN, LookUp, Patch, Refresh, RenameColumns, Sequence, ShowColumns, Shuffle, Sort, SortByColumns, Summarize, Table

Aggregate: Average, Max, Min, StdevP, VarP

Type conversion: AsType, Boolean, Dec2Hex, Decimal, Float, GUID, Hex2Dec, JSON, ParseJSON

Other: Blank, ColorFade, ColorValue, Error, Language, OptionSetInfo, RGBA, Trace, With

Available Templates

Templates are bundled with the plugin. Skills that use templates reference them via ${CLAUDE_SKILL_DIR}/../../templates/.

TemplateFilePattern
Greetingtemplates/topics/greeting.topic.mcs.ymlOnConversationStart welcome
Fallbacktemplates/topics/fallback.topic.mcs.ymlOnUnknownIntent with escalation
Arithmetictemplates/topics/arithmeticsum.topic.mcs.ymlInputs/outputs with computation
Question + Branchingtemplates/topics/question-topic.topic.mcs.ymlQuestion with ConditionGroup
Knowledge Searchtemplates/topics/search-topic.topic.mcs.ymlSearchAndSummarizeContent fallback
Custom Knowledge Sourcetemplates/topics/custom-knowledge-source.topic.mcs.ymlOnKnowledgeRequested with custom API (YAML-only)
Remove Citationstemplates/topics/remove-citations.topic.mcs.ymlOnGeneratedResponse citation stripping
Authenticationtemplates/topics/auth-topic.topic.mcs.ymlOnSignIn with OAuthInput
Error Handlertemplates/topics/error-handler.topic.mcs.ymlOnError with telemetry
Disambiguationtemplates/topics/disambiguation.topic.mcs.ymlOnSelectIntent flow
Agenttemplates/agents/agent.mcs.ymlGptComponentMetadata
Connector Action (generic)templates/actions/connector-action.mcs.ymlTaskDialog with connector (structural reference)
Knowledge (Public Website)templates/knowledge/public-website.knowledge.mcs.ymlPublicSiteSearchSource
Knowledge (SharePoint)templates/knowledge/sharepoint.knowledge.mcs.ymlSharePointSearchSource
Global Variabletemplates/variables/global-variable.variable.mcs.ymlGlobalVariableComponent

Frequently asked questions

What does the Int Reference AI skill do?

Reference tables for Copilot Studio YAML authoring: triggers, actions, variables, entities, Power Fx functions, templates. Preloaded by author and advisor agents.

Why use Int Reference on TypingMind?

Because you install it once and use it with any model. Int Reference is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Int Reference in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/microsoft/skills-for-copilot-studio/tree/main/skills/int-reference. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Int Reference?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Int Reference?

As many as you like. As long as a model supports skills, you can use Int Reference with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Int Reference AI skill free?

Yes. It is published on GitHub by microsoft under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇