MCP Servers

MCP servers give your AI agents hands. Instead of only generating text, an agent connected to an MCP server can search a database, create a calendar event, read a Notion page, or update a CRM record - all by calling actions that the server exposes.

What is MCP?

The Model Context Protocol (MCP) is an open standard for connecting AI agents to outside tools and data. You can think of an MCP server as a plug: one end speaks a language your agent understands, and the other end is wired up to a real service like Notion, GitHub, a filesystem, or your own internal API. Once the plug is in, the actions on the far side become available to your agent automatically.

You do not need to know any code to use MCP. Most MCP servers are hosted by their vendor and only need a URL and (usually) an access token.

---

MCP vs. Composio vs. Webhooks

FormWise gives you a few different ways to connect external services. They overlap, but each is best at a different job.

• MCP servers - The right choice when a vendor (or your team) already runs an MCP server and you want the agent to decide which actions to call during a conversation. Best for read-heavy, discovery-style work like "look something up", "summarise this", or "draft a record".
• Composio - A managed library of pre-built integrations with popular SaaS apps. Reach for Composio when you want a polished, one-click connection to a specific app and you do not want to host anything yourself. See Composio.
• Webhooks and HTTP nodes - Use these when you have a single, well-defined API call to make at a specific step of your Flow. They run deterministically rather than being chosen by the agent. See Node Types.

Rule of thumb: if the agent should pick when to call the action, use MCP or Composio. If your Flow should always call the action at a specific step, use a webhook or HTTP node.

---

Adding an MCP server

1. Open the Flow where you want to use MCP actions.
2. Switch to the Integrations tab at the top of the Flow editor.
3. Click Add MCP Server.
4. Enter the Server URL the vendor gave you (for example, https://mcp.notion.com/mcp or https://mcp.example.com/sse).
5. Give the connection a short Label so your team can recognise it later.
6. Optionally add a Description describing what the server does.
7. Choose an Authentication method and provide any required credentials (see below).
8. Click Connect.

[Screenshot: The "Connect MCP Server" dialog with the Server URL, Label, Description, and Authentication fields filled in.]

If FormWise can reach the server it will discover every action the server exposes and switch the dialog into the connected view. If it cannot reach the server you will see an inline error explaining what went wrong.

Authentication methods

Different MCP servers require different credentials. FormWise supports:

• None - For servers that do not require authentication, usually internal or local servers.
• Access Token / API Key - The most common option. Paste the token the vendor issued; FormWise sends it as a bearer token on every request.
• Custom Headers - For servers that expect a specific header (for example, X-Api-Key). Add as many key/value pairs as you need.

Tokens are stored encrypted and are only ever sent server-to-server. You will not see the full value again after saving.

---

What happens once it's connected

After a successful connect, FormWise queries the server for the list of actions it provides. Each action shows up in the Available Tools table inside the MCP configuration dialog with:

• The action name (for example, search_pages, create_issue).
• A short description the server provides.
• A checkbox to enable or disable that specific action.
• An Approval dropdown - Never, Always, or First Use - so you can require a human-in-the-loop check on sensitive actions.

[Screenshot: The Available Tools table inside the MCP dialog, with a list of discovered actions and approval dropdowns.]

These actions are then surfaced inside every Agent node in the same Flow. Open any agent, scroll to the tools section, and pick which MCP actions that specific agent is allowed to call. Different agents in the same Flow can use different subsets of actions.

For a deeper walk-through of how agents use tools, see Configuring Agent Nodes.

---

Picking which actions to enable

The Available Tools table is where you control what your agent is allowed to do. A few patterns work well:

• Start small. Tick two or three actions you actually plan to use. You can always come back and enable more.
• Read first, write later. Enable read-only actions (search, fetch, list) early, and only turn on write actions once the agent is reliable.
• Use approval for writes. Anything that creates, updates, or deletes data should have its Approval set to Always or First Use.
• Open an action to see its schema. Click the chevron on any action row to see its description and full input schema. If the schema asks for fields your agent could not reasonably know, that is a sign the action is not a good fit.

[Screenshot: An action detail panel showing the description and the JSON input schema for a selected MCP action.]

You can also set a Global Approval policy from the top of the dialog - Per Tool (respect each action's setting), Always (force approval on every call), or Never (skip approval entirely). Use Always in production when you want a hard safety brake.

---

Worked example: a Notion research agent

Say you want to build a chatbot that can look things up in your team's Notion workspace.

1. Get the URL. Notion publishes an MCP endpoint at https://mcp.notion.com/mcp.
2. Get a token. Generate an integration token in Notion's settings and share the pages you want the agent to read with that integration.
3. Add the server. In FormWise, open your Flow, go to Integrations, click Add MCP Server, paste the URL, label it "Notion", choose Access Token, and paste your token.
4. Connect. FormWise discovers actions like search, fetch, and update_page.
5. Pick safe defaults. Tick search and fetch and leave update_page off - a read-only agent is a good place to start.
6. Wire it into an agent. Open the Agent node that handles the conversation. In its tools list, enable the search and fetch Notion actions.
7. Test it. Run the Flow and ask, "Find our onboarding doc and summarise the first section." The agent decides on its own to call search, then fetch, then write the summary.

You did not write any code. You did not hard-code when to call Notion. The agent figured that out from the conversation.

---

Approval strategies

For anything that writes data, sends a message, or spends money, set the action's Approval to Always (or First Use if you trust it after a sanity check).

When approval is required, the agent will pause mid-conversation and surface the proposed call - including the exact arguments it wants to use - so a human can approve or reject it before it runs. This is your safety net for destructive actions.

A good default: read actions on Never, write or send actions on Always, until you have run them enough times to trust them.

---

Common pitfalls

• Connection errors. Double-check the URL. Some servers expose /mcp, others use /sse. FormWise will try both transports for SSE-style URLs, but the path still has to match what the vendor documented.
• Empty tool list. If the server connects but returns no actions, the token probably does not have access to anything yet. For Notion, GitHub, and similar apps, you typically have to share specific resources (pages, repos, projects) with the integration before they become visible.
• Action does not appear in the agent. Make sure the action is ticked in the MCP dialog's Available Tools table, and enabled in the Agent node's tools list. Both layers have to be on.
• Version skew. Vendors occasionally rename or remove actions on their MCP servers. FormWise auto-refreshes the list when you open the dialog, so re-opening it is enough to pick up changes. Any previously-enabled actions that no longer exist will be dropped silently.
• Forgotten credentials. If a connection used to work and suddenly stops, the token has probably expired or been revoked. Open the MCP dialog, click Change next to Authentication, and paste a fresh token.

---

Tips

• Start with one server and a small set of read-only actions. Add more once the agent is behaving well.
• Use the action Description that the vendor provides to decide what is safe to leave enabled. If a description is unclear, leave the action off.
• Connect MCP servers per-Flow, not per-organisation. This keeps the agent's available actions tightly scoped to the job at hand.
• Pair MCP with Notebooks and web tools for an agent that can read your knowledge base, search the live web, and act on external systems all in one conversation.

---

Next steps