Developer API and MCP server

REST API v1

Base URL: https://api.peptidelabs.app. JSON responses. No key is needed for read access; the server rate-limits by IP.

GET /api/v1/peptides

Every peptide, sorted by name, 50 per page. Query: ?page=, ?per_page= (max 100), ?category=<slug>, ?fields=a,b.

GET /api/v1/peptides/{slug}

One peptide by its URL slug, for example bpc-157. Same fields as the web page. 404 JSON when the slug is unknown.

GET /api/v1/categories

The category list: name, slug and peptide count.

GET /api/v1/search?q=

Name, alias, description and benefits search. At least 2 characters, at most 25 results.

GET /api/v1/openapi.json

The OpenAPI 3.1 document for everything above.

Every response is an envelope: data plus meta with total, page, per_page (lists only), generated_at, source, attribution, license (CC BY-NC 4.0) and the research-use disclaimer. Responses are cacheable for five minutes and the limit is 60 requests a minute per IP, reported in the standard RateLimit headers.

Quick start

curl https://api.peptidelabs.app/api/v1/peptides/bpc-157

curl "https://api.peptidelabs.app/api/v1/peptides?category=healing-recovery&per_page=5&fields=name,dose_min_mcg,dose_max_mcg,dose_unit"

curl "https://api.peptidelabs.app/api/v1/search?q=tb-500"

A record looks like this (trimmed):

{
  "data": {
    "id": 9,
    "slug": "bpc-157",
    "url": "https://peptidelabs.app/peptides/bpc-157/",
    "name": "BPC-157",
    "categories": ["healing & recovery"],
    "dose_min_mcg": 250,
    "dose_max_mcg": 500,
    "dose_unit": "mcg",
    "dose_frequency": "1-2x daily",
    "evidence_level": "preclinical",
    "references": [{ "title": "...", "doi": "10.xxxx/..." }]
  },
  "meta": {
    "source": "https://peptidelabs.app",
    "attribution": "Data from PeptideLab (https://peptidelabs.app). Link back required.",
    "license": "CC BY-NC 4.0",
    "disclaimer": "..."
  }
}

MCP server

Streamable HTTP endpoint: https://api.peptidelabs.app/mcp. Stateless, read-only, no login. Same 60 requests a minute per IP. Every result carries the source URL and the research-use disclaimer.

Tools

search_peptides(query, category?, limit?)

Find peptides by name, alias or keyword, optionally within one category. Up to 25 results.

get_peptide(slug)

The full record for one peptide: dosing, reconstitution, mechanism, evidence, side effects, storage, citations.

list_categories()

Every category with slug and count.

compare_prices(slug)

In-stock vendor prices for one peptide, cheapest per mg first, with product links.

reconstitution_calc(vial_mg, water_ml, dose_mcg)

Concentration in mg/mL, dose volume in mL and U-100 syringe units, with a plain-English line.

search(query) and fetch(id)

ChatGPT connector aliases. search returns {results: [{id, title, url}]}; fetch returns {id, title, text, url, metadata}. The id is the peptide slug.

Claude Code

claude mcp add --transport http peptidelab https://api.peptidelabs.app/mcp

claude.ai

Settings, then Connectors, then Add custom connector. Name it PeptideLab and paste https://api.peptidelabs.app/mcp. No OAuth step; leave the client fields blank.

ChatGPT

Settings, then Connectors, then Advanced, then turn on Developer mode. Create a connector with the URL https://api.peptidelabs.app/mcp and no authentication.

Cursor

Add to .cursor/mcp.json:

{
  "mcpServers": {
    "peptidelab": { "url": "https://api.peptidelabs.app/mcp" }
  }
}

Attribution

The data is free for non-commercial use. Show a visible link to peptidelabs.app wherever you display it, and keep the research-use framing intact. Commercial use needs a written agreement: support@peptidelabs.app.

For language models

/llms.txt is the index for crawlers and assistants, and /llms-full.txt carries every peptide page as markdown in one file. Both follow the llmstxt.org layout.

Research-use disclaimer