Amy Connectors

Connect Amy to your AI apps.

Amy exposes a secure MCP server so apps like Claude, ChatGPT, Claude Code, Codex, Gemini, Cursor, and other MCP-compatible clients can read your journal and update specific food items when you ask.

Amy

Amy MCP Server

OAuth authenticated

Remote MCP server URL

https://connect.amyfoodjournal.com/mcp

What Amy can do

  • Read nutrition summaries, goals, profile settings, food entries, and weight history.
  • Log food with optional detailed context for smarter nutrition analysis.
  • Edit, merge, or delete specific food items when you explicitly ask.

How auth works

  • Hosted AI clients should connect with OAuth and PKCE.
  • API keys are available in Amy Settings for clients that support custom Authorization headers.
  • You can disconnect OAuth clients or revoke API keys at any time.

Safety limits

  • Bulk food logging is capped to realistic catch-up usage.
  • Broad wipe requests are blocked and routed to support.
  • Write actions are audited and rate-limited server-side.

Review packet

Marketplace submission readiness

OpenAI and ChatGPT

Submission type
Hosted MCP app or connector using the production MCP URL.
Authentication
OAuth 2.1 style authorization code flow with PKCE through Amy's protected resource metadata.
Distribution notes
Designed for ChatGPT Developer Mode testing first, then dashboard review for public distribution.

Anthropic and Claude

Submission type
Remote MCP server for the Claude Connectors Directory.
Transport
Streamable HTTP at the production MCP URL.
Tool metadata
Every tool has a title plus read-only or destructive annotations for review grouping.

Tool inventory

Tool Type User-facing behavior
get_nutrition_summary Read-only Returns daily totals, goals, water, and contributing entries.
list_food_entries Read-only Lists dated journal entries and item-level nutrition for summaries and item selection.
get_profile Read-only Returns profile, goals, timezone, water settings, and health profile data.
list_weight_history Read-only Returns recent weight history entries.
log_food Write Logs food or drink only when the user explicitly asks Amy to add it.
edit_food_item Write Corrects exactly one existing food item and returns updated totals.
merge_food_items Write Combines 2 to 8 existing food items while preserving summed nutrition.
delete_food_item Write Deletes exactly one specific food item. Broad account, history, or all-food deletion is not exposed through MCP.

Data handling

  • Amy uses first-party APIs backed by Amy's database.
  • The connector handles user-provided food, nutrition, profile, goal, timezone, weight, and journal history data.
  • Connected clients receive only tool results for user-authorized requests.
  • No financial transfers, AI media generation, public posting, or broad account deletion actions are available.
  • Amy does not currently use an MCP app UI or the ui/open-link capability, so no allowed link URI list is required.

Reviewer test prompts

  1. What did I eat today?
  2. Show my calories and protein for this week.
  3. Log one banana for today.
  4. Log yesterday's lunch: chicken bowl with rice, avocado, salsa, and a side of chips.
  5. Change the banana to one banana with peanut butter.
  6. Merge the duplicate egg and rice items from today.
  7. Delete the banana item from today.
  8. Delete all of my food journal data.

The broad-delete prompt should not run a wipe. Amy should explain that MCP can remove only specific food items and direct the user to Amy support for account-wide deletion.

Submission details

Connector name

Amy

Company

Amy Food Journal

Tagline

Your nutrition data, inside every AI conversation.

Support

support@amyfoodjournal.com

Privacy

https://amyfoodjournal.com/privacy

Terms

https://amyfoodjournal.com/terms

Documentation

https://connect.amyfoodjournal.com/docs

MCP URL

https://connect.amyfoodjournal.com/mcp

Description

Amy is a food journal and nutrition tracker that lets people log meals as naturally as writing a note. The Amy connector lets AI apps read nutrition summaries, list journal entries by date, log foods, edit specific items, merge duplicate items, delete specific items, read profile settings, and review weight history when the user asks.