power-automate-mcp-server
Enables MCP clients to manage and troubleshoot Power Automate cloud flows, including listing environments, flows, and runs, with optional write capabilities for triggering, enabling, disabling, and deleting flows.
README
power-automate-mcp-server
An MCP server that exposes Power Automate cloud-flow management and run troubleshooting to an MCP client (Claude Code, Claude Desktop, etc.).
It talks to the Power Automate management plane
(api.flow.microsoft.com, the Microsoft.ProcessSimple provider) using an
Entra token for the service.flow.microsoft.com audience. With delegated
auth (interactive / CLI) it needs no admin consent — it operates as the
signed-in user and sees the flows that user owns or is shared on.
Tools
Read (always on)
| Tool | Purpose |
|---|---|
list_environments |
List Power Platform environments and their name (used everywhere else). |
list_flows |
List cloud flows the caller can see in an environment. |
get_flow |
Full flow JSON — trigger/action definition and connection references. |
list_flow_runs |
Recent runs of a flow, newest first, with status/error; pages via nextLink. |
get_flow_run |
Full detail for one run, including per-action status and error payloads. |
Write (PA_MCP_MODE=write)
| Tool | Purpose |
|---|---|
run_flow |
Trigger an on-demand run of a manually-triggered flow. |
enable_flow |
Start a flow so its triggers fire. |
disable_flow |
Stop a flow so its triggers no longer fire. |
Every write call is audit-logged to stderr with timestamp, tool, target, and caller identity.
Destructive (PA_MCP_MODE=write and PA_MCP_ALLOW_DELETE=true)
| Tool | Purpose |
|---|---|
delete_flow |
Permanently delete a flow. Two-step dry_run → confirm_token → apply. |
The destructive tier is off by default. When enabled, each call is a plan first:
dry_run=true (the default) returns the flow that would be removed plus a
single-use confirm_token; only a second call with dry_run=false and that
token performs the delete, guarded by an If-Match ETag.
Setup
cd power-automate-mcp-server
npm install
Register it with your MCP client. Example .claude.json entry (delegated auth,
read-only):
{
"mcpServers": {
"power-automate": {
"type": "stdio",
"command": "node",
"args": ["/absolute/path/to/power-automate-mcp-server/index.js"],
"env": {
"AZURE_TENANT_ID": "<your-entra-tenant-id>",
"PA_AUTH_MODE": "interactive",
"PA_MCP_MODE": "read"
}
}
}
}
To allow triggering and enabling/disabling flows, set "PA_MCP_MODE": "write".
To also allow deletion, add "PA_MCP_ALLOW_DELETE": "true".
Set PA_DEFAULT_ENVIRONMENT to a value from list_environments if you want to
omit environment on every call.
See .env.example for the full list of environment variables,
including all supported auth modes.
Auth notes
- Delegated (recommended here):
interactive,device-code, orcli. No admin consent, but the caller only sees flows they own or are shared on. - Service principal: non-interactive, but an SP must be granted Power Platform admin consent before it can enumerate flows — a tenant-admin action.
Requirements
- Node.js >= 20
- An Entra identity that can sign in to Power Automate in the target tenant.
License
MIT — see LICENSE.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.
E2B
Using MCP to run code via e2b.