MCP Power Automate

Local MCP server and Chromium extension for AI-operated Microsoft Power Automate flows.

The extension captures the browser session, current flow, snapshots, and token candidates automatically. The MCP server exposes one clean v1 tool surface so the AI can inspect, edit, validate, run, review, and revert flows without asking the user to click extension buttons just to make the connection work.

No Microsoft Entra ID app registration, admin consent, or custom enterprise app setup is required. The MCP uses your existing logged-in Chromium session.

What v1 Changes

• One command registry powers both MCP tools and HTTP bridge routes.
• The local bridge is single-owner; a second MCP instance fails fast if the bridge port is already occupied.
• Flow targeting is automatic-first through captured tabs, explicit connect_flow, catalog data, and snapshots.
• The extension is status and diagnostics only. It no longer requires manual refresh/sync buttons for normal operation.
• npm ships the built server and built extension together.

Install

Register the MCP:

codex mcp add power-automate-local -- npx -y @kaael1/mcp-power-automate

Find the packaged extension path:

npx -y @kaael1/mcp-power-automate extension-path

Load that folder in Chromium:

1. Open chrome://extensions or edge://extensions
2. Enable Developer Mode
3. Choose Load unpacked
4. Select the path printed by extension-path

Open or focus any Power Automate flow page. The extension captures the context automatically.

Check readiness:

npx -y @kaael1/mcp-power-automate doctor

Recommended AI Workflow

Ask your MCP client to:

1. doctor
2. get_context
3. connect_flow
4. get_flow
5. preview_flow_update
6. validate_flow
7. apply_flow_update
8. get_last_update

For run inspection and manual/request trigger tests, use list_runs, get_latest_run, get_run, get_run_actions, wait_for_run, get_trigger_callback_url, and invoke_trigger.

For Dataverse solution work, use list_solutions, list_solution_components, list_environment_variables, add_flow_to_solution, and create_flow_in_solution. Solution writes are intentionally limited to adding cloud flows to unmanaged solutions.

Public v1 Tools

• get_context
• doctor
• connect_flow
• list_flows
• list_solutions
• list_solution_components
• list_environment_variables
• add_flow_to_solution
• get_flow
• preview_flow_update
• validate_flow
• apply_flow_update
• get_last_update
• revert_last_update
• list_runs
• get_latest_run
• get_run
• get_run_actions
• wait_for_run
• get_trigger_callback_url
• invoke_trigger
• create_flow
• create_flow_in_solution
• clone_flow

Solutions And Environment Variables

The extension can capture Power Platform/BAP and Dataverse-audience tokens from the logged-in browser session. When those tokens are present, get_context and /health expose canManageSolutions and the MCP can inspect Dataverse solutions for the current or provided environment.

The solution-oriented surface supports inspection plus one focused write path for unmanaged solutions:

• list_solutions lists visible unmanaged solutions by default.
• list_solution_components lists components for a solution unique name, with optional enrichment for cloud flows and environment variables.
• list_environment_variables lists environment variable definitions and current values, optionally scoped to one solution.
• add_flow_to_solution adds an existing cloud flow as a workflow component in an unmanaged solution.
• create_flow_in_solution creates a blank cloud flow in the current environment, then adds it to an unmanaged solution.

Write operations such as creating solutions, setting environment variable values, publishing customizations, deletin

... View full README on GitHub