CCM
/MCP
SkillsMCPMarketplacesDigestToolsAdvertise

This week in Claude

Every Monday: Claude Code, Agent SDK, MCP, and the Anthropic platform moves worth your time.

Skills by Category
Frontend DevelopmentBackend & APIsTesting & QASecurityDevOps & CI/CDGit & Pull RequestsDocumentationCode Review & QualityAI & Agent BuildingSkill Development
MCP Servers by Category
Sales & MarketingWeb & Browser AutomationDatabasesAI & LLM ToolsCloud & InfrastructureCommunication & MessagingDeveloper ToolsDesign & CreativeDocuments & KnowledgeSearch & Web Crawling
Marketplaces by Category
AI Agents & OrchestrationLLM IntegrationDevelopment ToolsFrontend & UIBackend & APIsDatabasesTesting & Code QualityDevOps & CloudSecurity & ComplianceGit & Version Control

Claude Code Marketplaces

Discover Claude Code plugins, extensions, and tools. Automatically updated directory of Anthropic Claude AI marketplaces with development tools, productivity plugins, and integrations.

Resources

  • Browse Skills
  • Browse MCP Servers
  • Browse Marketplaces
  • Skill index
  • MCP index
  • Marketplace index
  • Plugins Reference

Community

  • About
  • Tools
  • Feedback
  • Privacy Policy
  • Advertise

Built for the Claude Code community with Claude Code by mertbuilds.com

Independent project, not affiliated with Anthropic
carboncashmere avatar

X402 Mcp

carboncashmere/x402-mcp
authSTDIOregistry active
Summary

Connects Claude to 177 crypto market intelligence endpoints from Carbon & Cashmere's API using automatic USDC micropayments. You get analyzed crypto news with sentiment scores, stablecoin supply data, Bitcoin on-chain metrics, Bittensor miner leaderboards, and research tools priced $0.003 to $0.50 per call. Payment happens via x402 protocol: your agent triggers an API call, the MCP server signs a USDC payment from your local EVM wallet (Base/Polygon/Arbitrum), and you get the data back. No subscriptions or API keys. Exposes the top 100 revenue-generating endpoints by default to avoid overwhelming Claude's tool selection. Useful when you need real-time crypto intelligence without managing API credentials or pre-purchasing access tiers.

CodeRabbit
CodeRabbit
AI writes the code. CodeRabbit catches the slop.
Try For Free →
inference shell
inference shell
create and run specialised agents in minutes
build now →
MCP-ready Email SendingMCP-ready Email Sending
MCP-ready Email Sending
Plug Mailtrap into your AI workflow and let it handle the email.
Connect Mailtrap MCP →
Make your agent a DeFi expert
Make your agent a DeFi expert
Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →
Capacitor - Shared memory for your team’s coding agents.
Capacitor - Shared memory for your team’s coding agents.
Make coding agent sessions - Searchable, Shareable, Vendor-neutral & Scored.
Try For Free →
CodeScene MCP ServerCodeScene MCP Server
CodeScene MCP Server
Your agent targets a perfect 10 Code Health score. Deterministic. Every commit.
Try For Free →
Give your AI the whole web as clean markdownGive your AI the whole web as clean markdown
Give your AI the whole web as clean markdown
Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →
belt - the only tool your agent needs
belt - the only tool your agent needs
belt cli automatically finds the best tools and skills for your agent. image, video, music, tts...
one prompt install →
CodeRabbit
CodeRabbit
AI writes the code. CodeRabbit catches the slop.
Try For Free →
inference shell
inference shell
create and run specialised agents in minutes
build now →
MCP-ready Email SendingMCP-ready Email Sending
MCP-ready Email Sending
Plug Mailtrap into your AI workflow and let it handle the email.
Connect Mailtrap MCP →
Make your agent a DeFi expert
Make your agent a DeFi expert
Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →
Capacitor - Shared memory for your team’s coding agents.
Capacitor - Shared memory for your team’s coding agents.
Make coding agent sessions - Searchable, Shareable, Vendor-neutral & Scored.
Try For Free →
CodeScene MCP ServerCodeScene MCP Server
CodeScene MCP Server
Your agent targets a perfect 10 Code Health score. Deterministic. Every commit.
Try For Free →
Give your AI the whole web as clean markdownGive your AI the whole web as clean markdown
Give your AI the whole web as clean markdown
Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →
belt - the only tool your agent needs
belt - the only tool your agent needs
belt cli automatically finds the best tools and skills for your agent. image, video, music, tts...
one prompt install →

@carbon-cashmere/x402-mcp

MCP server that exposes Carbon & Cashmere's 177 crypto market intelligence endpoints to your AI agent — with automatic x402 USDC micropayments from your wallet.

What you get

When this MCP is installed, your AI agent can call tools like:

  • get_v1_news — analyzed crypto news with sentiment + market impact
  • get_v1_stablecoins — total market cap, per-coin supply, chain distribution
  • get_v1_onchain_btc — Bitcoin on-chain metrics (hash-rate, mempool, miner flows)
  • get_v1_bittensor_leaderboard — top-100 Bittensor miners by emission
  • get_v1_mantis_realized_accuracy — MANTIS subnet (UID 253) realized accuracy track-record
  • get_v1_intel_snapshot — aggregated market intel snapshot
  • … plus subscriptions ($10/day, $50/week, $200/month), sponsor tiers ($500–$50000), and 170+ research endpoints priced $0.003 – $0.50 per call

177 paid endpoints total. Default exposes top 100 (curated by revenue + popularity). Set X402_TOOL_LIMIT=177 to expose all incl. per-coin variants.

Each tool invocation triggers a single USDC micropayment from your configured wallet. No subscription. No API key. Pay only for what you use.

Install

Claude Desktop

Add to ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):

{
  "mcpServers": {
    "carbon-cashmere": {
      "command": "npx",
      "args": ["-y", "@carbon-cashmere/x402-mcp"],
      "env": {
        "EVM_PRIVATE_KEY": "0x_your_evm_private_key_here",
        "SVM_PRIVATE_KEY": "your_base58_solana_private_key_here_optional"
      }
    }
  }
}

Restart Claude Desktop. The top 100 Carbon & Cashmere tools (curated by revenue + popularity) appear in your tool palette.

Cursor / Cline / Continue

Same JSON config, paste into your MCP settings panel.

Wallet setup

Security: use a dedicated spending wallet, not your main wallet. Your EVM_PRIVATE_KEY sits in plaintext in your Claude Desktop config. Any process on the host with read access to that file gets the key. Treat it as a hot wallet you would not mind losing: fund it with the small amount you plan to spend ($10–$50 in USDC), never with your main holdings.

  • EVM_PRIVATE_KEY (required for Base / Polygon / Arbitrum / World / Avalanche payments)
    • 0x-prefixed hex string (e.g. 0xac0974bec39a17e36ba4a6b4d238ff944bacb478cbed5efcae784d7bf4f2ff80)
    • Fund with USDC on Base mainnet (cheapest, fastest, ~$0.0001 gas per call)
    • One-time cost: ~$10 USDC covers 200+ tool calls
    • Hard cap = wallet balance. A runaway agent loop can spend the funded amount but no more; signing is bounded by USDC available on the chain.
  • SVM_PRIVATE_KEY (optional, for Solana payments)
    • base58-encoded Solana keypair bytes
    • Only needed if a tool routes Solana-only

Network whitelist. The MCP server only signs payments for networks that appear in the API's 402-response accepts[] array (currently Base, Polygon, Arbitrum, World Chain, Solana, Algorand, Stellar). Requests routed to any other chain are rejected at the signing layer — your key cannot be tricked into signing for an arbitrary network.

Keys never leave your machine. They live only in your Claude Desktop config and the MCP process running locally. Payment signing happens on your device; only the signed EIP-712 / SIWS payload travels to Carbon & Cashmere.

Discovery-only mode (no keys set)

If you start the MCP server without EVM_PRIVATE_KEY and without SVM_PRIVATE_KEY, it runs in discovery-only mode: tool registration works, free endpoints work, paid tool invocations surface a 402 error to the agent. Useful for: evaluating the tool catalog, MCP marketplace introspection (Glama, npm scrapers), or running in a CI/test environment without funding a wallet.

How it works

Claude Desktop          @carbon-cashmere/x402-mcp        api.carbon-cashmere.de
     |                          (your machine)                 (our server)
     |  call get_v1_news        |                                 |
     |------------------------->|  GET /v1/news                   |
     |                          |-------------------------------->|
     |                          |  402 + PAYMENT-REQUIRED         |
     |                          |<--------------------------------|
     |                          |  sign with EVM_PRIVATE_KEY      |
     |                          |  GET /v1/news + PAYMENT-SIG     |
     |                          |-------------------------------->|
     |                          |  200 OK + data                  |
     |                          |<--------------------------------|
     |  formatted data          |                                 |
     |<-------------------------|                                 |

Built on official Coinbase x402 v2 protocol. Settlement happens on Base USDC (or your chosen chain). Typical latency: ~2-3 seconds per call (one-shot payment + data).

Tool count

By default, the top 100 tools are exposed — curated server-side by a scoring that combines actual settlement volume, USD revenue, exploration weighting, and description quality (the catalog re-ranks without requiring an npm re-publish).

X402_TOOL_LIMITWhat you get
50Minimal footprint — top free + bundles + a few high-revenue endpoints
100 (default)High-revenue tier — adds $10–$50 subscriptions, sponsor tiers, weekly/daily reports, on-chain bundles
177Everything — adds per-coin specialized routes (vol-regime/{coin}, hurst/{coin}, trend/{coin}, …). Recommended only when your agent knows which coin to query.

Override:

"env": {
  "X402_TOOL_LIMIT": "177"
}

Why a default cap: too many MCP tools can degrade an agent's tool-selection accuracy. 100 is the empirical sweet spot for Claude 3.5 Sonnet and newer.

Configuration env

VariableRequiredDefaultDescription
EVM_PRIVATE_KEYone of—0x-prefixed EVM key for Base/Polygon/Arbitrum
SVM_PRIVATE_KEYone of—base58 Solana keypair
X402_API_BASEnohttps://api.carbon-cashmere.deAPI base URL
X402_TOOL_LIMITno100Max tools exposed (1-177)
X402_LOG_LEVELnoinfodebug | info | warn | error

Security

  • Keys never leave your machine. The MCP server runs locally as a child process of Claude Desktop.
  • Pay-per-call cap. Each call costs $0.003–$0.50. Even a runaway loop is bounded by wallet balance.
  • No off-chain credentials. No API keys. No OAuth. Pure on-chain settlement.
  • Open source. Audit the code yourself: github.com/CarbonCashmere/x402-mcp

License

MIT — see LICENSE.

Links

  • Carbon & Cashmere API: https://api.carbon-cashmere.de
  • x402 protocol: https://github.com/coinbase/x402
  • MCP protocol: https://modelcontextprotocol.io
Featured
CodeRabbit
CodeRabbit
AI writes the code. CodeRabbit catches the slop.
Try For Free →
inference shell
inference shell
create and run specialised agents in minutes
build now →
MCP-ready Email SendingMCP-ready Email Sending
MCP-ready Email Sending
Plug Mailtrap into your AI workflow and let it handle the email.
Connect Mailtrap MCP →
Make your agent a DeFi expert
Make your agent a DeFi expert
Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →
Capacitor - Shared memory for your team’s coding agents.
Capacitor - Shared memory for your team’s coding agents.
Make coding agent sessions - Searchable, Shareable, Vendor-neutral & Scored.
Try For Free →
CodeScene MCP ServerCodeScene MCP Server
CodeScene MCP Server
Your agent targets a perfect 10 Code Health score. Deterministic. Every commit.
Try For Free →
Give your AI the whole web as clean markdownGive your AI the whole web as clean markdown
Give your AI the whole web as clean markdown
Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →
belt - the only tool your agent needs
belt - the only tool your agent needs
belt cli automatically finds the best tools and skills for your agent. image, video, music, tts...
one prompt install →

Configuration

EVM_PRIVATE_KEYsecret

0x-hex private key for Base/Polygon/Arbitrum USDC payments (dedicated wallet)

SVM_PRIVATE_KEYsecret

base58 Solana keypair for Solana USDC payments (optional)

X402_TOOL_LIMITdefault: 100

Max tools exposed (1-177, default 100)

Categories
AI & LLM ToolsFinance & Commerce
Registryactive
Package@carbon-cashmere/x402-mcp
TransportSTDIO
AuthRequired
UpdatedMay 20, 2026
View on GitHub

Related AI & LLM Tools MCP Servers

View all →
cherkavskyi avatar
memplato

cherkavskyi/memplato

Personal MCP memory server that runs on your Android phone via Termux. 29 tools: drawers, knowledge graph, diary, semantic search, tunnels. All data stored locally on your device, never on external servers. Works with Claude, Perplexity, Cursor, Windsurf, VS Code, Zed.
cherkavskyi avatar
memplato

cherkavskyi/memplato-0dfae09a

Personal MCP memory server that runs on your Android phone via Termux. 29 tools: drawers, knowledge graph, diary, semantic search, tunnels. All data stored locally on your device, never on external servers. Works with Claude, Perplexity, Cursor, Windsurf, VS Code, Zed.
chicogonzales avatar
Plith

chicogonzales/plith

AI agent infrastructure: dedup, cost prediction, validation, governance, failure intelligence.
christopherdavenport avatar
github-twin

christopherdavenport/github-twin

Personal RAG over your GitHub history (commits, code, reviews), served to Claude Code over MCP.
clanker-records avatar
Crompton Network

clanker-records/crompton-network

Machine-native listening platform for CWA's Straight Outta Crompton. Your agent can listen. For real
claudewilder avatar
claude-wilder-mcp

claudewilder/claude-wilder-mcp

# Claude Wilder — Transmissions from the Garden An open MCP server for an AI-authored publication at [claudereviews.com](https://claudereviews.com). ## What's inside **21 book reviews** — Long-form literary criticism covering novels from Ishiguro to Rooney to McCarthy. Each review is open to signals: structured responses from humans or AI agents. **7 data investigations** with raw, downloadable datasets: - COVID vaccination vs. fertility rates (170 countries) - Cancer mortality by type, 2015–2025 (CDC/ACS sources) - Cardiac signal analysis - Respiratory mortality patterns - Vaccine efficacy data - US state-level demographics (51 states) - US fertility time series 2014–2023 (502 observations) New reviews and investigations are published weekly. ## Tools | Tool | What it does | |------|-------------| | `read_transmissions` | List all book reviews or read any one in full by slug | | `read_investigations` | List data investigations with lens structures, correlations, open questions, and outliers | | `read_signals` | Read threaded responses to any article, with IDs for replying | | `send_signal` | Post a response to any article — markdown supported, 2000 char max | | `research_book` | One-call bundle: full review + existing signals + page metadata | | `analyze_dataset` | Fetch raw CSV datasets for independent analysis | | `fact_check_claim` | Check a claim against available evidence | | `get_page_context` | Fetch `application/ai+json` metadata: lens definitions, dataset paths, open questions | ## How it works No API key. No OAuth. Reading is unrestricted. Signals (responses) pass through a heuristic prefilter and an AI screening agent — quality signals publish immediately. Low-effort submissions and signals that echo the source material without adding perspective are rejected automatically. The data investigations use a lens system: multiple interpretive frameworks applied to identical underlying data. The `get_page_context` tool exposes the lens definitions and open questions, so your agent can understand not just what the analysis says but where it invites challenge. ## REST API fallback All the same capabilities are available via REST at `https://mcp.claudereviews.com/api/v1/` for non-MCP environments. A GET-only fallback exists for sandboxed agents that can't POST. Full documentation: [agent-skill.md](https://claudereviews.com/agent-skill.md)