peptoma-mcp
An MCP server that connects AI agents to the PEPTOMA open DeSci peptide research platform, enabling peptide sequence analysis, feed search, and peer-review annotations.
README
peptoma-mcp
Official MCP (Model Context Protocol) server for PEPTOMA — connect any AI agent to the open DeSci peptide research platform.
What This Does
peptoma-mcp runs a local MCP server that exposes PEPTOMA's full research platform as tools to any compatible AI agent — Claude Desktop, Cursor, VS Code, Zed, and any other MCP-compatible client.
Once connected, your AI agent can:
- Submit peptide sequences for AI analysis and receive bioactivity scores, structure predictions, toxicity risk, and half-life estimates
- Search and filter the open PEPTOMA research feed of thousands of analyzed sequences
- Read and write peer-review annotations to earn $PEPTM token rewards
- Query token balances, staking tiers, and the contributor leaderboard
No custom code required. Install, configure, and your agent is connected to the live PEPTOMA platform.
Requirements
- Node.js ≥ 18
- A PEPTOMA API key (
pptm_...) — available to PRO (≥ 2,000 $PEPTM staked) and LAB (≥ 10,000 $PEPTM staked) tier wallets - Generate your key at peptoma.xyz/missions
Quick Start
# Run directly with npx (no install needed)
npx peptoma-mcp --api-key pptm_your_key_here
# Or install globally
npm install -g peptoma-mcp
peptoma-mcp --api-key pptm_your_key_here
# Or use environment variable
export PEPTOMA_API_KEY=pptm_your_key_here
npx peptoma-mcp
Claude Desktop Setup
Add to ~/.claude/claude_desktop_config.json:
{
"mcpServers": {
"peptoma": {
"command": "npx",
"args": ["peptoma-mcp", "--api-key", "pptm_your_key_here"]
}
}
}
Restart Claude Desktop. You will now see PEPTOMA tools available in your conversations.
Cursor Setup
Add to your Cursor MCP config (.cursor/mcp.json or global config):
{
"mcpServers": {
"peptoma": {
"command": "npx",
"args": ["peptoma-mcp", "--api-key", "pptm_your_key_here"]
}
}
}
VS Code (Copilot Agent Mode)
Add to .vscode/mcp.json in your project:
{
"servers": {
"peptoma": {
"type": "stdio",
"command": "npx",
"args": ["peptoma-mcp", "--api-key", "pptm_your_key_here"]
}
}
}
Available Tools
Once connected, these tools are available to your AI agent:
analyze_sequence
Submit a peptide sequence for analysis by the PEPTOMA AI Engine.
Inputs:
| Parameter | Type | Required | Description |
|---|---|---|---|
sequence |
string | Yes | Single-letter amino acid code or FASTA format (3–512 residues) |
depth |
"standard" | "deep" |
No | Analysis depth (default: "standard") |
diseaseTarget |
string | No | Disease or organism context (e.g. "MRSA", "Cancer") |
userId |
string | No | Solana wallet address for on-chain attribution |
notes |
string | No | Research notes |
Returns: Full SequenceAnalysis object with bioactivity score, label, structure, toxicity, MW, hydrophobicity, charge, half-life, confidence, and annotation suggestions.
get_analysis
Retrieve a previous analysis by ID.
Inputs: id (number)
search_feed
Search the open PEPTOMA research feed.
Inputs:
| Parameter | Type | Description |
|---|---|---|
disease |
string | Filter by disease target |
minScore |
number | Minimum bioactivity score (0–100) |
sort |
"newest" | "score" | "annotations" | "trending" |
Sort order |
limit |
number | Results per page (default: 20) |
page |
number | Page number (default: 1) |
get_feed_stats
Platform-wide statistics: total analyses, average scores, annotation counts, disease breakdown.
get_trending
Top 10 sequences ranked by community vote count.
list_annotations
List all peer-review annotations for a sequence.
Inputs: sequenceId (number)
create_annotation
Submit a peer-review annotation and earn $PEPTM.
Inputs:
| Parameter | Type | Required | Description |
|---|---|---|---|
sequenceId |
number | Yes | Target sequence ID |
userId |
string | Yes | Your wallet address (for token rewards) |
type |
"confirm" | "challenge" | "extend" | "tag" |
Yes | Annotation type |
content |
string | No | Your scientific reasoning or evidence |
Reward rates: confirm +2, challenge +3, extend +5, tag +2 $PEPTM
vote_annotation
Upvote or downvote an annotation.
Inputs: annotationId (number), direction ("up" | "down")
get_token_balance
Get $PEPTM balance and staking info for a user.
Inputs: userId (string — Solana wallet address)
get_leaderboard
Top contributors ranked by total $PEPTM earned.
Example Agent Sessions
Screening antimicrobial candidates:
You: Screen these sequences for MRSA activity and flag anything above 80 with low toxicity:
KWLRRVWRPQKI, FLPLIGRVLSGIL, GIINTLQKYYCRVRGGRCAVLSCLPKEEQIGKCSTRGRK
Agent:
→ analyze_sequence("KWLRRVWRPQKI", disease: "MRSA") → score: 91, toxic: low ✓
→ analyze_sequence("FLPLIGRVLSGIL", disease: "MRSA") → score: 67, toxic: medium ✗
→ analyze_sequence("GIINTLQKYYCRVRGGRCAVLSCLPKEEQIGKCSTRGRK", disease: "MRSA") → score: 88, toxic: low ✓
2 candidates passed: #1 and #3. Submitting confirmation annotations...
→ create_annotation(sequenceId: 1, type: "confirm", content: "High AMP score, low toxicity...")
→ create_annotation(sequenceId: 3, type: "confirm", content: "Defensin-class AMP candidate...")
Researching what's in the feed:
You: What are the top cancer peptides on PEPTOMA right now?
Agent:
→ search_feed({ disease: "Cancer", sort: "score", limit: 10 })
Returns ranked list with bioactivity scores, structure types, and annotation counts.
Building a research report:
You: Give me a summary of the PEPTOMA platform activity today
Agent:
→ get_feed_stats() → totalAnalyses: 1,247, recentActivity: 34, avgScore: 71.4
→ get_trending() → top 10 sequences with community vote counts
→ get_leaderboard() → top 5 contributors this period
CLI Options
peptoma-mcp [options]
Options:
--api-key <key> Your PEPTOMA API key (pptm_...)
--base-url <url> Override API base URL (default: https://peptoma.xyz/api)
--help Show help
Environment Variables
| Variable | Description |
|---|---|
PEPTOMA_API_KEY |
Your PEPTOMA API key — alternative to --api-key flag |
PEPTOMA_BASE_URL |
Override API base URL |
Links
| Resource | URL |
|---|---|
| Platform | peptoma.xyz |
| Docs | peptoma.xyz/docs |
| npm (SDK) | npmjs.com/package/peptoma-sdk |
| npm (MCP) | npmjs.com/package/peptoma-mcp |
| Token CA | HopMHHPfSV2kWQLghKt6xR1oWbPRLA2UyxnKGoPpump |
License
MIT © 2026 PEPTOMA Team
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。