Your AI is guessing about stocks.
Connect Distilla to Claude or ChatGPT and ask in plain English. Building your own agent? The same data is available through the Tools API.
7-day free trial. Works in Claude, ChatGPT, Claude Code, Cursor, WorkBuddy, Doubao, and more.
Pick one to see the prompt, what the agent pulls, and what comes back.
For PMs and active investors
Prompt
Scan every stock that moved more than 5% in the last trading session across the US, Japan, Korea, Hong Kong, and China. Explain each move and group them into themes.
Web search alone
Headlines for a handful of famous tickers. No record of why the rest moved.
With Distilla, the agent queries
What came back · Sep 25, 2026 session
| Ticker | Move | Why it moved | Theme |
|---|---|---|---|
| DELL | +5.0% | Raised FY27 AI-server revenue guide to $74B from $60B | AI infrastructure |
| CRDO | +7.7% | DustPhotonics deal adds silicon photonics to its AI connectivity stack | AI infrastructure |
| 4716.T | +8.1% | Oracle Japan cloud revenue +32%; full-year guide signals AI-led growth | AI infrastructure |
| 2338.HK | −6.7% | Ex-dividend, plus HK power-equipment selloff after an Oracle data-center payment delay | AI infra wobble |
| ZS | −10.1% | Chief revenue officer exit, after a sector-wide AI-security rally | Cybersecurity unwind |
| 6301.T | +6.0% | Komatsu wins ~C$1.5B, 300+ machine mining order with autonomous haulage | Company-specific |
26 stocks moved 5% or more, grouped into 5 themes. 7 of 13 decliners were cybersecurity names giving back a sector rally. Every explanation comes from price_explanation.
Also try
Synthesize this earnings season across hundreds of reports. Which patterns repeat across sectors?
Review the past earnings season for HK-listed tech stocks. Any common themes?
Same knowledge layer, same JSON. Pick the one that fits how you work.
01 — for analysts and PMs
Connect it to the assistant you already use and ask in plain language. Nothing to build.
02 — for developers and quants
The same tools over HTTP, for dashboards, screens, backtests and anything you'd schedule.
In their own words.
With Claude + Distilla, it takes 30 seconds to synthesize 360 views and what's important across sell-side and podcasts.
I use Workbuddy and Distilla to build screening and analysis routines that get me and my followers great stock ideas. Very impressed.
Before Distilla, it was impossible for me to vibe-code a robust and complete dashboard for my 100+ stock universe. Distilla's pre-indexed price attribution, earnings and company facts make this efficient and accurate.
I use Distilla on Claude to research stocks for my personal investment and to prepare for my hedge fund interviews. Super high returns on a small monthly subscription.
One point-in-time, citable database. Tag names are the entities and tools your agent sees.
438K+
Price-move explanations
Significant daily stock moves, each with what drove it.
7M+
Qualitative knowledge cells
Segments, guidance, KPIs, and topic changes extracted from filings and calls.
180K+
Sell-side research reports
Summarized and tagged by company and industry in the Public Library.
11K+
Podcast episodes
Transcribed, summarized, and tagged by company and industry.
5 markets
US · China · Japan · Hong Kong · Korea
~50K
Company events added weekly, across 180+ event categories.
10K+
Sell-side actions added weekly.
Point-in-time
Every fact is dated and traceable to its source document.
Business drivers and products
Company-by-company drivers and peer categories for bottom-up screens.
Filings, transcripts, and news
The source documents behind every fact.
Financials and consensus
Historical statements and analyst estimates by fiscal period.
Prices and valuation
Daily prices and multiples such as P/E and EV/EBITDA.
Add a card to start a 7-day free trial of Distilla MCP, the Tools API, and Distilla Lite. You will not be charged until the trial ends, and you can cancel any time before then.
Each time your AI looks something up in Distilla, like Adobe's last 12 earnings reactions or Sony's segment results, that is one tool call. A quick question uses a few. A deep analysis like a DCF can use 10 to 20.
Distilla's research in your browser, with no AI required. Look up any covered company for a one-page tearsheet, earnings history, every big stock move with its reason, valuation ranges and peers, plus the research and podcast library.
Open LiteDistilla is a data and research tool, not investment advice.
Can't find your answer? Email the team.
Add the plugin and start a 7-day free trial.
https://api.distilla.ai/mcp