- name
- web3tech
- description
- Crypto project research, new coin discovery, due diligence, developer analysis, whitepaper validation, code similarity checks, and AI-powered development monitoring — all via the web3tech MCP server.
- version
- 1.0.0
- metadata
- openclaw
- requires
- env
- primaryEnv
- WEB3TECH_API_KEY
- emoji
- 🔬
- homepage
- https://web3tech.org
Web3Tech Crypto Research
Use this skill for crypto project research with the web3tech MCP server.
Before You Start
- Check whether the
web3techMCP server is available. - If unavailable, say so briefly and continue with a best-effort fallback.
- Do not invent missing metrics, rankings, timelines, whitepaper facts, or developer history.
Choose the Right Path
| User wants | Start with |
|---|---|
| Project due diligence, comparison, technical risk | references/due-diligence.md |
| New coin discovery, novelty screening, watchlist | references/new-coin-screening.md |
| Output formatting | references/output-templates.md |
Tool Rules
web3tech_search_coins— resolve naming ambiguity only.web3tech_coin_analysis— default baseline for any serious research.web3tech_top_analyzed_coins— discover high-quality projects with existing scores.web3tech_hot_coins— find trending projects by views and community activity.web3tech_new_coins— source candidates, not final judgments.web3tech_new_coin_detail— review before branching into deeper diligence.web3tech_search_developers— find developers by name or GitHub handle.web3tech_project_developers— list before profiling individuals.web3tech_developer_profile— only for contributors that materially affect the answer.web3tech_project_timeline— when recent execution quality matters.web3tech_supervision_overview— AI daily development health check.web3tech_code_similarity— when originality, fork risk, or code reuse matters.web3tech_whitepaper_analysis— when claims, feasibility, or tokenomics matter.web3tech_knowledge_search— interpret technical concepts, not replace project evidence.
Efficiency
- Due diligence: 3–6 tool calls.
- New coin screening: 3–8 tool calls.
- Do not run deep tools on every candidate from a broad search.
- Stop once you have enough evidence for a defensible answer.
Output Rules
- Evidence-first conclusions.
- Separate observed facts from interpretation.
- If evidence is mixed or thin, say so explicitly.
- Treat novelty as a lead, not proof.
- Do not present early-stage projects as validated winners.