A professional-grade crypto research agent for deep-diving EVM and Solana tokens with automated social sentiment analysis and alerting.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install token-research
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install token-research using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Token Research?
The Token Research skill is a high-performance analysis tool designed for AI agents to evaluate cryptocurrency tokens across major chains including Base, Ethereum, Arbitrum, and Solana. It bridges technical on-chain data with real-world social sentiment by integrating DexScreener, GoPlus security audits, and advanced X/Twitter searches.
By utilizing Openclaw Skills, developers can automate the identification of high-potential projects while filtering out low-quality meme coins through rigorous narrative assessment and developer verification. The skill offers two distinct modes: a deep_research path for exhaustive project vetting and a shallow_dive path for quick technical snapshots.
Token Research Use Cases
- Automated monitoring of new token launches on EVM chains and Solana.
- Deep-dive technical and social sentiment analysis for specific contract addresses.
- Building automated watchlists with tiered risk assessments (Tier 1 to Tier 3).
- Real-time alerting for high-signal WATCH or APE rated tokens via Telegram, Discord, or automated phone calls.
- Batch processing multiple tokens simultaneously using parallel sub-agents to maximize efficiency.
How Token Research Works
- Data Retrieval: The skill calls the DexScreener API to check real-time price action, liquidity depth, and volume metrics.
- Security Audit: It utilizes GoPlus to perform contract security checks, identifying potential honeypots or high-risk permissions.
- Social Verification: The agent performs advanced X/Twitter searches to verify developer claims, check KOL mentions, and gauge organic community growth.
- Narrative Assessment: A specialized scoring phase evaluates the project's uniqueness, viral potential, and product-market fit.
- Risk Synthesis: The agent combines all data points into a final verdict, triggering alerts for high-conviction signals and archiving the report in the file system.
Token Research Setup
To integrate this research capability into your workflow, ensure your environment has the necessary API keys for Twitter/X and the required shell scripts for alerting. Openclaw Skills requires a structured directory for output.
# Initialize the ape-call script for emergency alerts
chmod +x ./scripts/ape-call.sh
# Set your Twitter API Key
export TWITTERAPI_KEY='your_api_key_here'
Ensure that the reports/ and watchlists/ directories are present in your workspace to enable persistent data logging.
Token Research Data Schema & Taxonomy
The skill organizes its findings into a standardized Markdown-based data schema for easy retrieval and auditing:
| Component | Format | Storage Path |
|---|---|---|
| Deep Research Reports | Markdown File | reports/YYYY-MM-DD/[report-name].md |
| Token Watchlist | Appended Markdown | watchlists/YYYY-MM/watchlist.md |
| Batch Analysis | Summary File | reports/YYYY-MM-DD/[N]-token-analysis.md |
Each watchlist entry captures the Ticker, Chain, Contract Address (CA), Entry Market Cap, Current Market Cap, and a specific Narrative Catalyst.
Token Research Advanced Features
- Parallel sub-agent spawning for concurrent research on 5+ token batches.
- Mandatory Ape-Call integration to ensure no high-signal opportunities are missed by the owner.
- Cross-referencing logic that searches DEV_HANDLE history to verify authentic endorsements and filter out fake hype.
- Tiered risk categorization (Tier 1, 2, and 3) to help manage portfolio exposure based on signal strength.
- Automated narrative scoring including heat levels from Strong (Novel concept) to Weak (Generic/Repetitive).
SKILL.md
Loading
Related Openclaw Skills
AgentWallet
A non-custodial smart wallet framework for AI agents featuring on-chain spending limits and passkey-based human control.
Playwright Pro
A production-grade testing toolkit for AI coding agents to generate, fix, and manage resilient Playwright test suites.
Multimedia to Obsidian
Transform multimedia documents into intelligent, searchable Obsidian notes using multimodal AI analysis.
Solopreneur Business Management
A local-first solo business operating system for tracking revenue, sales pipelines, and weekly strategic reviews.
X/Twitter Research Skill
An advanced research tool that enables AI agents to perform deep-dive analysis of X/Twitter trends, user activity, and niche community conversations.
XHS-AutoContentByHot
An automated content creation workflow that transforms trending Baidu topics into complete XiaoHongShu posts with AI-generated copy and images.
Featured*
