Builds a pure-markdown investment research vault in Karpathy's LLM Wiki style by interviewing you and turning YOUR analysis framework into a living CLAUDE.md.
Pull the machine-readable Bigdata.com (RavenPack) substrate — financials, estimates, prices, daily sentiment, annotated chunks — through the official SDK and /v1 REST endpoints with built-in cost guards.
Pulls market and historical financial data for US-listed companies from yfinance and outputs one standardized JSON ready for valuation work.
One-command installer, credential configurator, and health doctor for the 19 official Gangtise (岗底斯) investment-research OpenAPI skills.
Unpacks your Google Takeout Gemini export, classifies every conversation, and produces a structured insight report.
Aggregates Chinese A-share news flashes, regulatory announcements and retail forum sentiment into structured JSON or Markdown.
Pulls live quotes for 20 core China A-share pharma stocks, ranks 7 sub-sectors with gainer/loser TOP 3 and money-flow estimates, then pushes a rich-text daily report to Feishu.
Runs an adversarial teardown of a founder, KOL, company, or product you envy — separating marketing bubble from real signal and mapping the validated playbook onto your own resources.
Drives a disciplined discover-inspect-plan-query-validate loop over ktx-connected databases so analytical answers are correct, not just runnable.
Turns your claude-mem observation history into focused, queryable knowledge bases you can chat with.
Lets Claude query PrimeKG to explore links between genes, drugs, diseases and phenotypes.
Search scientific papers and pull structured full-text experimental data through the BGPT MCP server.
Runs a full French statutory audit of annual accounts across 7 NEP-based phases and produces a structured audit opinion report.