Skip to content

microsoft/power-platform-skills / plugins/code-apps/skills/add-mcscopilot

add-mcscopilot

Verified

Adds Microsoft Copilot Studio connector to a Power Apps code app. Use when invoking Copilot Studio agents, sending prompts to agents, or integrating agent responses.

πŸ“‹ Shared Instructions: shared-instructions.md - Cross-cutting concerns.

Add Microsoft Copilot Studio

Workflow

  1. Check Memory Bank β†’ 2. Add Connector β†’ 3. Configure β†’ 4. Build β†’ 5. Update Memory Bank

Step 1: Check Memory Bank

Check for memory-bank.md per shared-instructions.md.

Step 2: Add Connector

Get the connection as described in connector-reference.md.

CLI 1.3.0+ (PA_INLINE_CONNECTIONS=yes): tell the user that adding Microsoft Copilot Studio may create a connection with their account, then run the add without -c and handle the result with the outcome table in connector-reference.md:

pa app add data-source --connector microsoftcopilotstudio --non-interactive

Older CLIs: run /list-connections to find the Microsoft Copilot Studio connection ID (if none exists, create one with pa connection create as described in connector-reference.md), then:

pa app add data-source --connector microsoftcopilotstudio -c <connection-id>

Step 3: Configure

Ask the user which Copilot Studio agent they want to invoke and what operations they need.

Agent Setup Prerequisites (manual steps the user must complete in Copilot Studio):

  1. Publish the agent: In Copilot Studio, click Channels β†’ select Teams β†’ add to Teams β†’ click Publish.
  2. Get the agent name: Under Channels, click "Web app". The connection string URL contains the agent name. Example: https://...api.powerplatform.com/copilotstudio/dataverse-backed/authenticated/bots/cr3e1_myAgent/conversations?... β€” the agent name is cr3e1_myAgent.

ExecuteCopilotAsyncV2 -- execute an agent and wait for the response:

Use the ExecuteCopilotAsyncV2 operation (path: /proactivecopilot/executeAsyncV2). This is the only endpoint that reliably returns agent responses synchronously. It is the same endpoint used by Power Automate's "Execute Agent and wait" action.

const result = await MicrosoftCopilotStudioService.ExecuteCopilotAsyncV2({
  message: "Your prompt or data here", // Can be a JSON string
  notificationUrl: "https://notificationurlplaceholder" // Required by API but unused; any URL works
});

// Response structure:
// result.responses β€” Array of response strings from the agent
// result.conversationId β€” The conversation ID
// result.lastResponse β€” The last response from the agent
// result.completed β€” Boolean indicating if the agent finished

Important: Agents often return responses as JSON strings. Parse the responses array to extract meaningful data:

const agentResponse = result.responses?.[0];
if (agentResponse) {
  const parsed = JSON.parse(agentResponse);
  // Extract specific fields, e.g., parsed.trend_summary
}

Use Grep to find specific methods in the generated service file (generated files can be very large β€” see connector-reference.md).

Known Issues

  • ExecuteCopilot (/execute) -- fire-and-forget, only returns ConversationId, not the actual response. Do NOT use this.
  • ExecuteCopilotAsync (/executeAsync) -- returns 502 "Cannot read server response" errors. Do NOT use this.
  • Conversation turn model (/conversations/{ConversationId}) -- only works after /execute, which doesn't provide responses. Do NOT use this.
  • Response casing varies -- check all variations: conversationId, ConversationId, conversationID.

Step 4: Build

npm run build

Fix TypeScript errors before proceeding. Do NOT deploy yet.

Step 5: Update Memory Bank

Update memory-bank.md with: connector added, agent name configured, configured operations, build status.

reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 469kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why β€” repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" β€” or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 321kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 321kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself β€” your human partner chose inline execution, or no subagent tool is availableStars: 321kUpdated Claude CodeCodexVerifiedInstall