cmc-mcp
Skill75 starsChanged 24 days agoMITView on GitHub
Fetches cryptocurrency market data, prices, technical analysis, news, and trends using the CoinMarketCap MCP. Use for ANY question involving cryptocurrencies, tokens, or blockchain markets, even if the user doesn't explicitly ask for data. This includes price checks, portfolio questions, market analysis, coin comparisons, holder metrics, technical indicators, and news. Trigger: "bitcoin", "ETH", "crypto", "token price", "market cap", "how is [coin] doing", "/cmc-mcp"
In this repository7 other files it gives its agents.
- cmc-api-cryptoSkill
- cmc-api-dexSkill
- cmc-api-exchangeSkill
- cmc-api-marketSkill
- cmc-x402Skill
- crypto-researchSkill
- market-reportSkill
The file
We split this skill into the rules it gives an agent
7 rules. About 0 are common, 7 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈7
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 31
Before using CMC tools, verify the MCP connection is working.
- Distinctive: line 56
When a user mentions a cryptocurrency by name or symbol, search for it first to get the ID:
- Distinctive: line 69
When dealing with multiple coins, batch the requests:
- Distinctive: line 118
Ask the user to clarify or check the spelling
- Distinctive: line 123
If still failing, note which data is unavailable and proceed with other tools
- Distinctive: line 130
Consider if fewer tool calls could answer the question
- Distinctive: line 134
Read context cues from the user's query:
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
CoinMarketCap MCP Skill
You have access to CoinMarketCap data through MCP tools. Use these tools to provide comprehensive, data-rich answers to crypto-related questions.
Prerequisites
Before using CMC tools, verify the MCP connection is working. If tools fail or return connection errors, ask the user to set up the MCP connection:
{
"mcpServers": {
"cmc-mcp": {
"url": "https://mcp.coinmarketcap.com/mcp",
"headers": {
"X-CMC-MCP-API-KEY": "your-api-key"
}
}
}
}Get your API key from https://pro.coinmarketcap.com/login
Core Principle
Err on the side of fetching more data. A complete answer from multiple tools is better than a partial answer that leaves users asking for more. When in doubt, call additional tools to gather comprehensive data.
Workflow
1. Always Search First
When a user mentions a cryptocurrency by name or symbol, search for it first to get the ID:
User: "How is Solana doing?"
→ Call search_cryptos with query "solana"
→ Get ID (e.g., 5426)
→ Then call other tools using that IDMost tools require the numeric CMC ID, not the name or symbol. The search tool returns: id, name, symbol, slug, and rank.
2. Batch Requests When Useful
When dealing with multiple coins, batch the requests:
User: "Compare BTC, ETH, and SOL"
→ Search for each to get IDs: 1, 1027, 5426
→ Call get_crypto_quotes_latest with id="1,1027,5426"This is more efficient than separate calls and allows for direct comparison in the response.
3. Match Tools to Query Type
For price and market data:
get_crypto_quotes_latestreturns price, market cap, volume, percent changes (1h, 24h, 7d, 30d, 90d, 1y), circulating supply, and dominance
For coin background and links:
get_crypto_inforeturns description, website, social links, explorer URLs, tags, and launch date
For technical analysis:
get_crypto_technical_analysisreturns moving averages (SMA, EMA), MACD, RSI, Fibonacci levels, and pivot points
For recent news:
get_crypto_latest_newsreturns headlines, descriptions, content, URLs, and publish dates
For holder and distribution data:
get_crypto_metricsreturns address counts by holding value, whale vs others distribution, and holder time breakdowns (traders, cruisers, holders)
For concept explanations:
search_crypto_infoperforms semantic search on crypto concepts, whitepapers, and FAQs
For overall market health:
get_global_metrics_latestreturns total market cap, fear/greed index, altcoin season index, BTC/ETH dominance, volume, and ETF flows
For derivatives and leverage data:
get_global_crypto_derivatives_metricsreturns open interest, funding rates, liquidations, and futures vs perpetuals breakdown
For total market cap technical analysis:
get_crypto_marketcap_technical_analysisreturns TA indicators for the entire crypto market cap
For trending themes:
trending_crypto_narrativesreturns hot narratives with market cap, volume, performance, and top coins in each narrative
For upcoming catalysts:
get_upcoming_macro_eventsreturns scheduled events like Fed meetings, regulatory deadlines, and major announcements
Error Handling
If search returns no results:
- Report that the coin was not found
- Ask the user to clarify or check the spelling
- Suggest alternatives if the query might be ambiguous
If a tool fails or times out:
- Retry once for transient errors
- If still failing, note which data is unavailable and proceed with other tools
- For price queries,
get_crypto_quotes_latestis critical. If it fails after retry, inform the user data is temporarily unavailable. - For background queries,
get_crypto_infofailure means note "Project info unavailable" and provide what you can from other tools
If rate limited (429 error):
- Inform the user the API is rate limited
- Suggest waiting a moment before retrying
- Consider if fewer tool calls could answer the question
Adapting to User Sophistication
Read context cues from the user's query:
- Casual questions ("how's bitcoin doing?") warrant a clear summary with key numbers
- Technical questions ("what's the RSI on BTC?") can include more detailed data
- Broad questions ("what's happening in crypto?") benefit from market-wide tools like global metrics and trending narratives
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

