{
  "markdown": "# Heathy — nutrition data (MCP server)\n\nNutrition for **917 whole plant foods** and **39 Ayurvedic herbs**, reconciled across\n**11 national food-composition datasets** — including **IFCT 2017 (India)**, which\nisn't available as an API anywhere else.\n\n**Endpoint:** `https://mcp.heathy.org/mcp` — Streamable HTTP (JSON-RPC 2.0)\n**Website:** <https://heathy.org> · **Contact:** hello@heathy.org\n**Free while in preview.**\n\n> This repository contains **documentation only** — the manifest and these notes.\n> The server is hosted; there is nothing to install.\n\n---\n\n## What makes it different\n\nMost nutrition APIs hand back a single number from a single national table. This one\nreturns the number *and its provenance*:\n\n- the **median** across every dataset that has the food,\n- **how many** sources contributed and **which** ones,\n- the **min–max spread** between them.\n\nSo an agent can cite a figure instead of asserting it, and can tell when the sources\ndisagree. Guava vitamin C, for example, comes back as `228 mg` from **10 sources**\nwith a range of `99.2–273` — a spread worth knowing about before stating it as fact.\n\n**Indian foods are the standout.** IFCT 2017 isn't queryable as an API elsewhere, so\nquestions about karela, horse gram or millets get real data rather than a US\nsubstitute.\n\n## Connecting\n\nAdd to any MCP client that supports remote servers:\n\n```json\n{\n  \"mcpServers\": {\n    \"heathy\": {\n      \"type\": \"streamable-http\",\n      \"url\": \"https://mcp.heathy.org/mcp\"\n    }\n  }\n}\n```\n\nNo API key, no signup. A service card describing the server is at\n<https://mcp.heathy.org/>.\n\n## Tools\n\n### `lookup_food`\n\nFull per-100 g nutrient profile for a whole plant food, with provenance.\n\n| Argument | Type | |\n|---|---|---|\n| `food` | string | **required** — slug, common name, or regional alias |\n| `nutrients` | string[] | optional — limit to specific nutrients rather than all ~168 |\n\n### `search_foods`\n\nResolve a name to a slug, including Hindi, Sanskrit and botanical aliases. Call this\nfirst when you have a plain-English or local name.\n\n| Argument | Type | |\n|---|---|---|\n| `query` | string | **required** |\n| `kind` | `food` \\| `herb` \\| `any` | optional, default `any` |\n| `limit` | integer | optional, 1–50, default 10 |\n\n### `nutrient_ranking`\n\nRank foods by a nutrient per 100 g — the \"highest in iron\" question.\n\n| Argument | Type | |\n|---|---|---|\n| `nutrient` | string | **required** — slug or name |\n| `limit` | integer | optional, 1–100, default 20 |\n| `category` | string | optional — e.g. `fruit`, `legume`, `millet` |\n| `min_sources` | integer | optional — exclude thinly-sourced outliers |\n\n### `herb_monograph`\n\nAn Ayurvedic herb monograph. Traditional use and modern research are kept **strictly\nseparate** — `traditional` records what the tradition claims, `modern` records what\nresearch shows with an evidence tier. **Every response carries a mandatory `safety`\nblock** (cautions, pregnancy guidance, drug interactions).\n\n| Argument | Type | |\n|---|---|---|\n| `herb` | string | **required** — English, Sanskrit, Hindi or botanical name |\n\n## Examples\n\n| Call | Result |\n|---|---|\n| `search_foods(\"karela\")` | `gourd-karela-raw` |\n| `lookup_food(\"guava\", [\"vitamin-c\"])` | 228 mg · 10 sources · range 99.2–273 |\n| `nutrient_ranking(\"iron\", category: \"legume\")` | horse gram 8.76 mg · soya bean 8.29 mg |\n| `herb_monograph(\"haldi\")` | turmeric — traditional, research, safety |\n\n## Response shape\n\nEvery tool returns the same envelope:\n\n```jsonc\n{\n  \"data\":        { /* the payload */ },\n  \"attribution\": [\"USDA FoodData Central …\", \"IFCT 2017, ICMR–NIN …\"],\n  \"source_url\":  \"https://heathy.org/food/guava-flesh-only-raw/\",\n  \"disclaimer\":  \"Informational only — not medical advice.\",\n  \"license\":     \"https://heathy.org/terms/\"\n}\n```\n\n**Please reproduce `attribution` and `disclaimer` when you use a response, and cite\n`source_url`.** For herbs, surface the `safety` block alongside any information you\npass on.\n\n## Data sources\n\nUSDA FoodData Central · UK CoFID · **IFCT 2017 (ICMR–NIN, India)** · ANSES-Ciqual\n(France) · BLS (Germany) · Frida (Denmark) · Matvaretabellen (Norway) · MEXT (Japan)\n· TACO (Brazil) · FAO/INFOODS WAFCT (West Africa).\n\nFull credits and licence terms: <https://heathy.org/sources/>\n\n## Licence and terms\n\nRepository documentation is released under\n[CC BY-SA 4.0](https://creativecommons.org/licenses/by-sa/4.0/). Composition figures\nare open government data, credited per response and at the link above. The\nreconciliation, recipes, herb monographs and original content are © Heathy — see\n<https://heathy.org/terms/>. Use of the hosted service at mcp.heathy.org is governed\nby those terms.\n\n**Not medical advice.** The data is informational and does not diagnose or treat\nanything. A 100% raw-vegan diet does not supply adequate vitamin B12 or vitamin D.\n",
  "bytes": 4823,
  "sha": "0266dc62b3ce806ebcbfb5802ef63cbcd1ca59da0594a13c374bf2037aa10a79",
  "repo_slug": "chandata/heathy-mcp",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_org_heathy_nutrition_1e780339/readme"
}