$npx -y skills add github/awesome-copilot --skill flowstudio-power-automate-mcpFoundation skill for Power Automate via FlowStudio MCP — auth setup, the reusable MCP helper (Python + Node.js), tool discovery via list_skills / tool_search, and oversized-response handling. Load this skill first when connecting an agent to Power Automate. For specialized wo
| 1 | # Power Automate via FlowStudio MCP — Foundation |
| 2 | |
| 3 | This skill is the **plumbing layer**. It gives an AI agent a reliable way to |
| 4 | talk to a FlowStudio MCP server, discover what tools are available, and handle |
| 5 | the responses cleanly. The actual workflow narratives live in four specialized |
| 6 | skills that all build on this one. |
| 7 | |
| 8 | > **Real debugging examples**: [Expression error in child flow](https://github.com/ninihen1/power-automate-mcp-skills/blob/main/examples/fix-expression-error.md) | |
| 9 | > [Data entry, not a flow bug](https://github.com/ninihen1/power-automate-mcp-skills/blob/main/examples/data-not-flow.md) | |
| 10 | > [Null value crashes child flow](https://github.com/ninihen1/power-automate-mcp-skills/blob/main/examples/null-child-flow.md) |
| 11 | |
| 12 | > **Requires:** A [FlowStudio](https://mcp.flowstudio.app) MCP subscription (or |
| 13 | > compatible Power Automate MCP server). You will need: |
| 14 | > - MCP endpoint: `https://mcp.flowstudio.app/mcp` (same for all subscribers) |
| 15 | > - API key / JWT token (`x-api-key` header — NOT Bearer) |
| 16 | > - Power Platform environment name (e.g. `Default-<tenant-guid>`) |
| 17 | |
| 18 | --- |
| 19 | |
| 20 | ## Which Skill to Use When |
| 21 | |
| 22 | Skills are organized by **use-case intent**, not by which tools they call. |
| 23 | Multiple skills reuse the same underlying tools — pick by what the user is |
| 24 | trying to accomplish. |
| 25 | |
| 26 | | The user wants to… | Load this skill | |
| 27 | |---|---| |
| 28 | | Make or change a flow (build new, modify existing, fix a bug, deploy) | **`flowstudio-power-automate-build`** | |
| 29 | | Diagnose why a flow failed (root cause analysis on a failing run) | **`flowstudio-power-automate-debug`** | |
| 30 | | See tenant-wide flow health, failure rates, asset inventory | **`flowstudio-power-automate-monitoring`** *(Pro+)* | |
| 31 | | Tag, audit, classify, score, or offboard flows | **`flowstudio-power-automate-governance`** *(Pro+)* | |
| 32 | | Just connect, set up auth, write the helper, parse responses | this skill (foundation) | |
| 33 | |
| 34 | **Same tools, different lenses.** `flowstudio-power-automate-build` and `flowstudio-power-automate-debug` |
| 35 | both call `update_live_flow`, `get_live_flow`, and the run-error tools — they |
| 36 | differ in *direction* (forward vs backward) and *intent* (compose vs diagnose). |
| 37 | `flowstudio-power-automate-monitoring` and `flowstudio-power-automate-governance` both call the Store |
| 38 | tools — they differ in *audience* (ops vs compliance) and *outcome* (read |
| 39 | health vs write metadata). Don't try to memorize "which tools belong to which |
| 40 | skill"; pick the skill by what the user is doing. |
| 41 | |
| 42 | --- |
| 43 | |
| 44 | ## Source of Truth |
| 45 | |
| 46 | | Priority | Source | Covers | |
| 47 | |----------|--------|--------| |
| 48 | | 1 | **Real API response** | Always trust what the server actually returns | |
| 49 | | 2 | **`tool_search` / `list_skills`** | Authoritative tool schemas, parameter names, types, required flags | |
| 50 | | 3 | **SKILL docs & reference files** | Workflow narrative, response shapes, non-obvious behaviors | |
| 51 | |
| 52 | If documentation disagrees with a real API response, the API wins. Tool schemas |
| 53 | in this skill (or any other) may lag the server — call `tool_search` to confirm |
| 54 | the current shape before invoking a tool you haven't used recently. |
| 55 | |
| 56 | --- |
| 57 | |
| 58 | ## How Agents Discover Tools |
| 59 | |
| 60 | The FlowStudio MCP server (v1.1.5+) exposes two **non-billable** meta-tools that |
| 61 | let an agent load only the tools relevant to the current task. Use these in |
| 62 | preference to `tools/list` (which loads all 30+ schemas at once) or guessing |
| 63 | tool names. |
| 64 | |
| 65 | | Meta-tool | When to call | |
| 66 | |---|---| |
| 67 | | `list_skills` | Cold start — see the available bundles (`build-flow`, `create-flow`, `debug-flow`, `monitor-flow`, `discover`, `governance`) and pick one | |
| 68 | | `tool_search` with `query: "skill:<name>"` | Load the full schema set for one bundle (e.g. `skill:debug-flow`) | |
| 69 | | `tool_search` with `query: "select:tool1,tool2"` | Load specific tools by name (e.g. when chaining across bundles) | |
| 70 | | `tool_search` with `query: "<keywords>"` | Free-text search when the user request is ambiguous (e.g. `"cancel run"`) | |
| 71 | |
| 72 | The server's `tool_search` bundles are intentionally **narrower than this |
| 73 | skill family** — they're starter packs of the most-likely-needed tools per |
| 74 | intent. A workflow skill (e.g. `flowstudio-power-automate-debug`) may pull a bundle and |
| 75 | then call `tool_search` again for additional tools as the workflow progresses. |
| 76 | |
| 77 | ```python |
| 78 | # Cold start — pick a bundle by intent |
| 79 | skills = mcp("list_skills", |