Sign in Book a demo
Back to Tools
Productivity & Docs API Key
Build Your Agent View Forkable Docs
forkable logo

Forkable MCP Tools for AI Agents

Let an agent list this week's meals, browse menus, check dietary restrictions, pick a replacement, or skip a delivery.

Available Tools

7 tools available in this integration

Forkable.CheckMealRestrictions

Check whether picking this item (with the given modifier selections) would

Forkable.ListMenuItems

Fetch the full menu (sections, items, prices, modifiers) for one or more menu IDs.

Forkable.ListMyMeals

List your upcoming Forkable delivery days and what's currently picked for each.

Forkable.ListMyMealsThisWeek

List your upcoming Forkable delivery days and what's currently picked for each.

Forkable.PickMeal

Replace your assigned meal on one delivery with a different menu item.

Forkable.SetMeal

Replace your assigned meal on one delivery with a different menu item.

Forkable.SkipMeal

Skip a day — removes your assigned meal from a Forkable order.

Why agent-optimized tools matter

Built for intent, not API schemas

Ask the agent to "show my meals this week, find a menu item that fits my restrictions, and replace Tuesday's selection." Arcade picks the right Forkable actions without guessing menu or delivery identifiers.

Up to 100x fewer tokens

Forkable reads return structured delivery, menu, price, and modifier details, so the agent can compare meals and check restrictions without loading an unstructured menu or full account history into the model.

Complete the workflow end to end

List this week's meals, browse menus, check restrictions, replace a selection, and skip a delivery day through one toolkit designed around the complete meal-planning task.

Make actions predictable and controlled

Arcade tools return structured results and useful errors across users and runs. Teams can apply scope restrictions, redaction, and approval policies so agents follow the same controls every time.

Securing Forkable

The more capabilities agents get, the more useful they are. However, more capabilities bring more risks of undesired effects such as data loss and exfiltration.

This is why Arcade provides you with contextual rules that dictate how and when a tool is allowed to run. Arcade’s contextual access lets you govern tool visibility and behavior at every stage of execution and control who can see a tool, what inputs are allowed, and what comes back.

Learn about Contextual Access Policies

For Forkable, you can implement some of these policies:

Access when tools are listed for a user
  • Hide Forkable.SkipMeal from agents unless they operate under the roles your IDP marks as meal-plan owners, so a routine agent cannot remove an upcoming delivery.
  • Restrict Forkable.PickMeal to agents operating under your approved meal-planning role, so unreviewed automation cannot replace a person's assigned meal.
  • Hide Forkable.SetMeal from agents outside the roles your IDP marks as meal-plan owners, so ordinary workflows cannot alter a subscriber's scheduled meal.
Pre-execution before each tool call
  • Require human approval before Forkable.PickMeal changes an assigned meal, so an agent cannot select an unintended dish or modifier.
  • Hold Forkable.SkipMeal until a human approves the delivery date, so an agent cannot remove a meal that someone still expects.
  • Require human approval before Forkable.SetMeal replaces an assigned meal, so an agent cannot commit a choice without a person confirming it.
Post-execution after the tool returns
  • Redact the restriction names and dietary explanations returned by Forkable.CheckMealRestrictions before they reach agents that do not need health-related meal information.
  • Deny Forkable.ListMenuItems results to agents outside your approved menu-planning role, so unrelated agents cannot inspect your available meals, prices, or modifiers.
  • Redact the selected meal names and delivery-day details from Forkable.ListMyMealsThisWeek results for agents without the meal-planning role, so personal food choices are not exposed to unrelated workflows.
  • Strip selected meal names from Forkable.ListMyMeals results for agents without the meal-planning role, so unrelated workflows see delivery dates without a person's food choices.

How to use Forkable with your agent

Connect Forkable to an MCP client and configure your shared credential once. Follow the setup from authentication to your first meal lookup, then try the example prompts below.

Your questions, answered

A Forkable MCP server exposes meal-planning capabilities as structured tools an AI agent can call through the Model Context Protocol. The agent can review upcoming deliveries, browse menus, check meal restrictions, change selections, and skip a delivery day.

Forkable does not provide a separate MCP server in this integration. Arcade hosts the Forkable integration and adds authenticated tool access, approval policies, redaction, and one connection path for supported clients.

Arcade agents can list this week's deliveries and selected meals, browse menu sections and items with prices and modifiers, check restrictions for a selected item, replace a meal, and skip a delivery day.

Forkable uses one shared API key or secret configured for the integration. That credential determines which Forkable account and actions the agent can access; there are no per-user OAuth scopes in this setup.

No. Users do not each connect their own Forkable account through OAuth. The integration uses one credential configured once, which Arcade brokers rather than exposing to the model.

Yes. Contextual Access policies can require human approval before changing a meal or skipping a delivery day. The call proceeds only after the policy's conditions have been satisfied.

Arcade proxies Forkable API calls and does not persist your content as part of normal tool execution. The shared credential is brokered and stored encrypted at rest, and the model never receives it.

Any framework or client that can connect to Arcade's MCP gateway or tool APIs works, including LangChain, OpenAI Agents, CrewAI, Claude Desktop, Claude Code, Cursor, Visual Studio Code, Microsoft Copilot Studio, and GitHub Copilot.

Arcade refreshes access while the configured authorization remains valid. If the Forkable credential is revoked, rotated, or can no longer be used, calls stop until the credential is reconnected.

Forkable's own rate limits and quotas apply to calls made through Arcade. Arcade does not remove those limits, so retry behavior should follow the limits and response guidance returned by Forkable.

Ready to actually ship your agent?

Connect your agent to Forkable with a shared API key, agent-optimized tools and approval controls for consequential actions.

Build Your Agent Read the Forkable Quickstart