Skip to main content
Glama
Sign In
PepBank Peptide Research MCP

PepBank Peptide Research MCP

Official
by pepbank
OverviewSchemaSourceRelated ServersDiscussionsAdmin
JavaScript
Hybrid

PepBank Peptide Research MCP: Studies, Clinical Trials and Cited Findings

Look up what studies found about more than 160 peptides, such as BPC-157, semaglutide and GHK-Cu. Each result links to its paper and says if it came from people, animals or cells.

PepBank MCP connects your AI to PepBank, a database of peptide research. Add it to Claude, ChatGPT, Cursor or another MCP app. Your AI can then find a peptide by name and read a summary of its evidence. It can also search study findings and clinical trials, and compare up to five peptides. Each finding links to the sentence in the paper that supports it and cites its PMID or DOI. Developer guide: pepbank.io/developers.

Server address

https://pepbank.io/mcp

Transport: Streamable HTTP. The server runs at this address. This repository holds its listing files. The server code is not in it.

Related MCP server: Hormonaly MCP Server

Add it to your AI app

Claude: open Customize, then Connectors. Select Add custom connector. Name it PepBank and enter https://pepbank.io/mcp.

ChatGPT: turn on Developer mode in Settings, under Security and login. Open Plugins and select +, then Add custom MCP server. Name it PepBank and enter https://pepbank.io/mcp.

Grok: open grok.com/connectors. Select New Connector, then Custom. Enter https://pepbank.io/mcp.

Claude Code:

claude mcp add --transport http pepbank https://pepbank.io/mcp

Codex (config.toml):

[mcp_servers.pepbank]
url = "https://pepbank.io/mcp"

Cursor (~/.cursor/mcp.json):

{ "mcpServers": { "pepbank": { "url": "https://pepbank.io/mcp" } } }

VS Code (.vscode/mcp.json):

{ "servers": { "pepbank": { "type": "http", "url": "https://pepbank.io/mcp" } } }

Gemini CLI (settings.json):

{ "mcpServers": { "pepbank": { "httpUrl": "https://pepbank.io/mcp" } } }

Other apps: add https://pepbank.io/mcp as a remote server with Streamable HTTP.

Command line and local server (npm)

The pepbank package runs the same server over stdio for apps that need a local server, and answers from the terminal:

npx pepbank brief BPC-157
npx pepbank search tendon healing
npx pepbank trials GHK-Cu

Local server config for MCP apps:

{ "mcpServers": { "pepbank": { "command": "npx", "args": ["-y", "pepbank"] } } }

Tools

Tool

What it does

resolve_peptide

Finds the peptide that a name, alias or code means

get_peptide_brief

Reads the evidence summary for one peptide, with key findings, trials, studied doses and approval status

get_community_practice

Reads what community sources report about how people use one peptide: uses, doses, side effects and cautions, each with its sources. Trials did not test these reports

search_peptides

Lists peptides and filters them by name, category or evidence level

search_findings

Searches study findings, with the source sentence and the PMID or DOI for each

search_publications

Finds papers and shows which ones have checked findings

search_trials

Finds registered clinical trials, with status, phase and posted results. The list view shows up to 100 trials per page

search_live_sources

Searches Europe PMC for papers that PepBank does not cover yet. These results are not checked

get_record

Reads one full peptide, paper, finding or trial record

get_evidence_map

Counts the findings for one peptide by outcome, evidence class and direction

compare_peptides

Compares the evidence for 2 to 5 peptides

get_changes

Lists data changes since a data version

replay_query

Runs a saved search again on the data version it first used

get_freshness

Shows when each source was last imported

get_my_usage

Shows your daily limits and how much of them you used

Prompts and resources

Name

Type

What it does

evidence_summary

Prompt

Summarizes the evidence for one peptide, with sources and coverage limits

compare_evidence

Prompt

Compares the evidence for 2 to 5 peptides

pepbank://guide

Resource

Explains evidence classes, coverage, data versions, citations and limits

Plugin package

This repository is also a plugin package that connects the server and adds one skill, skills/pepbank-peptide-research/SKILL.md.

  • plugin.json and mcp.json: Agent Plugins format (Hermes)

  • .claude-plugin/plugin.json and .mcp.json: Claude Code plugin format

  • .cursor-plugin/plugin.json: Cursor plugin format

  • openclaw.plugin.json: OpenClaw (ClawHub)

Install the skill with the skills CLI:

npx skills add pepbank/peptide-research-mcp

Questions to try

  • What do studies say about BPC-157 for tendon healing?

  • Are there clinical trials testing GHK-Cu?

  • Compare the evidence for semaglutide and tirzepatide.

  • What doses do people report for BPC-157, and what doses did studies use?

Data and credit

PepBank data is under CC BY 4.0. When you show it, credit it as "Data from PepBank" with a link to pepbank.io. Each tool result gives a credit line to use. The files in this repository are under the MIT License.

Links

Mira Labs LLC runs PepBank. This is not medical advice.

This server cannot be deployed

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Maintenance

ActivityMaintained
ResponsivenessSyncing

Resources

Related MCP Connectors

Related MCP Servers

  • life-science-mcp

    achillesrasquinha
    A
    license
    A
    quality
    B
    maintenance
    Enables unified access to 110 life science APIs and databases, including genomics, proteomics, chemistry, literature, and clinical data. Users can query genes, proteins, compounds, pathways, and more through natural language.
    3
    MIT

  • PubTator 3 MCP

    pipeworx-io
    A
    license
    Not graded
    quality
    C
    maintenance
    Enables AI agents to resolve genes, diseases, chemicals, variants and species to normalized ids, search ~36M PubMed abstracts and PMC full texts by free text or entity, and retrieve the machine-extracted relations between them with the supporting sentences and PMIDs. Supports auditing individual relations with full evidence passages and pulling per-article entity annotations with character offsets.
    255 npm
    MIT