- Home
- Skills
- Microck
- Ordinary Claude Skills
- Crypto Research
crypto-research_skill
- Python
124
GitHub Stars
2
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill microck/ordinary-claude-skills --skill crypto-research- metadata.json867 B
- SKILL.md5.7 KB
Overview
This skill provides comprehensive cryptocurrency market research by coordinating specialized AI agents that run in parallel. It delivers multi-angle analysis including price checks, technical indicators, news sentiment, macro correlations, and investment opportunity suggestions. Use it to get fast, configurable snapshots or in-depth multi-agent reports on specific digital assets or the broader crypto market.
How this skill works
The system selects a research mode (lightweight, comprehensive, or output-only) based on the request, then launches a set of agents focused on market conditions, coin-level analysis, macro correlations, news sentiment, movers, and investment plays. Each agent gathers data from multiple sources, writes structured outputs to a timestamped directory, and the skill summarizes results with file locations and agent completion status. Agents accept parameters such as ticker symbols and model class to control depth and cost.
When to use it
- Quickly check current conditions or price action for a specific coin (BTC, ETH, SOL, etc.).
- Request a comprehensive research report covering technical, fundamental, news, and macro signals.
- Investigate investment opportunities or trade ideas across market segments.
- Assess news-driven sentiment and recent developments affecting a token.
- Run background, automated analysis with file-only outputs for downstream workflows.
Best practices
- Start with lightweight (haiku) mode for quick answers, then upgrade to comprehensive for deeper research.
- Specify tickers and timeframes explicitly to focus agents and reduce noise.
- Check timestamps on outputs to verify data freshness before acting on insights.
- Review all agent outputs—different agents catch different signals and edge cases.
- Retry or rerun individual agents if an agent times out; partial outputs are preserved.
Example use cases
- User asks "What's happening with Bitcoin?" — run lightweight mode for a fast BTC snapshot.
- Portfolio manager requests a cross-asset macro correlation scan to inform allocation decisions.
- Trader wants top movers and sentiment for the last 24 hours — run market + movers + news agents.
- Investor seeks vetted investment plays across layer-1 tokens — run comprehensive multi-agent analysis.
- Automation: schedule nightly output-only runs to populate a timestamped research archive.
FAQ
Three model classes control depth and cost: haiku (fast, cheap), sonnet (balanced), and opus (deep, high-quality). Modes map to agent sets: lightweight (haiku agents), comprehensive (all agents), and output-only (files only).
How do I specify which coin to analyze?
Provide a ticker symbol (e.g., BTC, ETH, SOL); the coin-analyzer defaults to BTC if none is given.