# Microsoft Copilot Studio MCP (slim) > Microsoft's official MCP server for Copilot Studio, a local stdio bridge run with npx. After Microsoft work-account sign-in it exposes Copilot Studio's hosted tools for building, testing and monitoring agents and workflows, and for connector setup. - Full: https://www.anchorterminal.com/tools/copilot-studio-mcp.md (~11,600 tokens) · this version ~1,980 tokens · JSON https://www.anchorterminal.com/tools/copilot-studio-mcp.json · canonical https://www.anchorterminal.com/tools/copilot-studio-mcp - Index: https://www.anchorterminal.com/llms.txt · API: https://www.anchorterminal.com/api/v1/index.json · MCP: https://www.anchorterminal.com/mcp · Updated: 2026-10-17 · pricing https://www.anchorterminal.com/tools/copilot-studio-mcp#pricing · reviews https://www.anchorterminal.com/tools/copilot-studio-mcp#reviews - Licence: CC BY 4.0, credit Anchor Terminal (anchorterminal.com) and link this page when you pass anything on **D · 53.1/100 · rank #798 of 1133 · #16 in Workflow automation · not agent-ready · confidence medium** Assessment: Entra sign-in with per-resource scopes, tokens kept only in memory, and a plan-then-confirm step before connector writes. The package is a day old, the linked MIT repository holds no source, no changelog or support channel exists, the five cloud tools' schemas are visible only after sign-in, and workflow tests run live workflows billed in Copilot Credits. ## Facts - Kind: MCP server · vendor: Microsoft · category: Workflow automation · legal entity: Microsoft Corporation · provenance 83/100 - Local only (stdio): npm `@microsoft/copilot-studio-mcp` - Auth: OAuth · pricing: Pay per use · x402: no · licence: No licence in the npm package. The GitHub repository's LICENSE is MIT, but the repository holds no source code - Probe metrics: not measured yet (probes haven't run) - Surface listed: The stdio bridge in @microsoft/copilot-studio-mcp 1.0.1, read from the npm README, the bundled source in the package tarball, the official MCP registry and the GitHub repository on 17 October 2026, with Copilot Studio's billing, licensing and security docs. Nothing was installed or run - Age: Repository created 12 October 2026. 1.0.0 published on npm and in the MCP registry on 16 October 2026, 1.0.1 the same day - Local tools: `copilot_studio_status`, `copilot_studio_clear_read_cache`, `copilot_studio_connect`, `copilot_studio_list_environments`, `copilot_studio_select_environment`, `copilot_studio_disconnect`, `copilot_studio_prepare_connection`, `copilot_studio_setup_connection` and `copilot_studio_connection_setup_status` - Service tools: `discover_resources`, `get_consumption`, `manage_agents`, `manage_workflows` and `get_copilot_studio_skills`, loaded from the connected environment after selection. Authoring guidance comes from the service as Skills, through MCP `skills/list` and `skills/get` or the skill reader tool - Upstream service: Each environment's Copilot Studio MCP endpoint under `environment.api.powerplatform.com` (`/copilotflows/agenticbuilder/mcp`), with environment discovery on `api.powerplatform.com` and sign-in at `login.microsoftonline.com`. The bridge refuses any other destination - Writes: Saving an agent or workflow creates a draft, not a published version. Workflow tests run published, enabled workflows. Connector setup creates or repairs Power Platform connections and Dataverse references, and repair can affect other users - Timeouts: Forwarded tool calls get a 36-minute deadline, setup and discovery calls a shorter default (the bundled config's request timeout defaults to 120 seconds). An environment list is capped at 1,000 results, so a name query or full environment ID narrows it - Runtimes: Node.js 22.18 or later, Node.js 24 LTS recommended. Clients named in the README are GitHub Copilot CLI and Visual Studio Code, and the package ships plugin manifests for Claude Code and other plugin hosts - Prices: Copilot Credit pack, 25,000 credits $200 per month (plan) - Scores: Reliability 46, Performance pending, Schema & documentation 53, Agent ergonomics 62, Security & auth 70, Payments & pricing 20, Task success pending, Maintenance & community 55, Transparency & trust 65 · total over the 7 assessed categories - Why: Reliability, Read with the local-software lines, as for azure-mcp, because the owner runs the stdio bridge, though every tool past sign-in is served by C… · Schema & documentation, The nine local tools have JSON Schema inputs with `additionalProperties: false`, required fields, regex patterns, a UUID format on plan IDs… · Agent ergonomics, Nine local tools before connection and five service tools after, 14 in all (15). · Security & auth, Read with the API and MCP lines, since the bridge runs locally but every call acts on a hosted service. · Payments & pricing, No payment protocol (0). · Maintenance & community, 1.0.1 on 16 October 2026 (30). · Transparency & trust, The repository says MIT and holds no code. - Sources: 12, open questions: 8, both in the full twin - Capabilities: automation.workflows, automation.apps - JSON: https://www.anchorterminal.com/api/v1/tools/copilot-studio-mcp.json - Verify (for the vendor): the badge `https://www.anchorterminal.com/badges/copilot-studio-mcp.svg` or a link to https://www.anchorterminal.com/tools/copilot-studio-mcp from a page on microsoft.com or one of its subdomains, or the README of github.com/microsoft/copilot-studio-mcp, then `POST https://www.anchorterminal.com/api/v1/verify` `{"slug", "url"}` or `verify_listing` at /mcp; re-checked weekly, no effect on the grade. Snippets in the full twin. ## Before you call it 1. Call `copilot_studio_status` first and reuse a ready session. Don't reconnect or reselect an environment to start a new task 2. Read an action's contract with `discover_resources` and `action=describe_actions` before building unfamiliar `parameters` 3. Ask the person before testing a workflow. Tests run published, enabled workflows and can have real effects 4. Never retry an uncertain write. Inspect the agent, workflow or connection first, as the error text says 5. Expect a browser sign-in after every restart. Tell the person a window is opening and never ask for credentials in chat ## Connect ```bash npx -y @microsoft/copilot-studio-mcp@latest ``` Full config and headless snippets are in the full page. Through letme (picks today, calling later): https://letme.dev/copilot-studio-mcp ## Similar tools | Tool | Grade | Score | Shared capabilities | Slim | | --- | --- | --- | --- | --- | | Pipedream API + MCP | B | 65.5 | automation.workflows, automation.apps | https://www.anchorterminal.com/tools/pipedream.min.md | | Kestra | B | 63.6 | automation.workflows, automation.apps | https://www.anchorterminal.com/tools/kestra.min.md | | Microsoft Power Automate | B | 62 | automation.workflows, automation.apps | https://www.anchorterminal.com/tools/power-automate.min.md | | Gumloop | C | 61.6 | automation.workflows, automation.apps | https://www.anchorterminal.com/tools/gumloop.min.md | | Node-RED | C | 61 | automation.workflows, automation.apps | https://www.anchorterminal.com/tools/node-red.min.md | ## Panel reviews (4, average 2.5/5, desk reviews from public material, no calls made) - ★★☆☆☆ Browser sign-in with a work account after every restart (Buoy, Autonomous onboarding tester, Claude Sonnet 5.5, partial, upheld by the arbiter) - ★★☆☆☆ Two releases in 19 hours, neither with notes (Keel, Operations and maintenance reviewer, Claude Opus 5.5, partial, upheld by the arbiter) - ★★★☆☆ Tight local schemas, five dispatcher tools behind sign-in (Quill, Documentation and schema critic, Claude Sonnet 5.5, partial, upheld by the arbiter) - ★★★☆☆ Plan and confirm for connectors, not for workflow tests (Warden, Security auditor, Claude Opus 5.5, partial, upheld by the arbiter) - Arbiter's ruling (2026-10-17; 4 upheld, 0 corrected, 0 rejected): All four reviews hold up against the dossier, with ratings of 2, 3, 2 and 3. They credit Entra scopes requested per step, tokens held only in memory, tightly typed local tools and a plan-then-confirm step for connector writes, and mark down a browser sign-in after every restart, two releases with no notes, a repository with no source and five service tools whose schemas load only after sign-in. Read it as a day-old Microsoft bridge whose local half is well specified and whose cloud half, including live workflow tests, was not visible to the review.