Stop your AI from guessing about the markets. Kaleidoscope MCP feeds Claude, ChatGPT, Gemini, and any MCP client real, cited filing data — point-in-time accurate and verifiable down to the source.
MCP is an open standard — bring it to the assistant you already use
Frontier LLMs are brilliant at language and reasoning. But three structural gaps — that no amount of model scaling fixes — mean Claude alone can't be trusted with serious research. Kaleidoscope MCP fills exactly that layer.
The latest 10-K, this morning's M&A announcement, last quarter's 13F — none of it exists for a model frozen at its training date.
With MCP: Claude queries filings, news (2020–2026, refreshed daily), and institutional holdings as they're published — and cites them.
Ask Claude to rebuild a company's pipeline as of Q3 2018 and it returns today's pipeline (wrong) or a confident reconstruction (untrustworthy).
With MCP: Every tool accepts an as_of date and returns the state of the world as it was knowable then — leak-safe by construction.
Training is dominated by the companies and drugs that got written about. The failures rarely make the corpus, so any base rate Claude computes is inflated.
With MCP: Base rates are computed over a survivorship-complete population — failures, delistings, and discontinued programs included in the denominator.
One natural-language question. Claude loads the right tools and returns a synthesized, cited read — here it is working inside Claude.
Our research desk puts Kaleidoscope MCP to work — activism, institutional flow, and where smart money is quietly rotating. Every claim traced to a filing.
A founder presses for resolution at lululemon, Ackman nears the point of no return at Howard Hughes, and the broader flow picture shows smart money rotating out of mega-cap tech into energy and power.
Request access and we'll get you a bearer token. Currently free for approved users — no credit card required.