Your personal trading analyst, on call.
The Alphalog AI workspace is a founder-grade agent with live access to your journal, your research, and the market. Ask anything — from "why did I lose money last week?" to "screen for oversold mid-caps with improving fundamentals."
How it works
Step 01
Alphalog AI sees your data
With your permission, the agent reads your journal, open positions, bias tags, and research notes — so every answer is specific to your book, not generic advice.
Step 02
Live market tools
Quotes, fundamentals, news, options chains, and charts are all available as agent tools. You ask in English; Alphalog fetches what it needs.
Step 03
Usage you can see
Token usage is tracked and visible, with tier-appropriate limits. No surprise bills. No "pay for another seat."
Journal-aware conversation
It already knows your book.
Alphalog AI starts every conversation with full context on your recent trades, open positions, and bias patterns. No copy-paste. No re-explaining who you are.
- Reads your journal, portfolio, and bias tags
- Cites specific trades when it answers
- Respects per-user privacy — your data never trains shared models
Market-connected
Real tools, real data, real answers.
The agent calls real APIs: Polygon quotes, FMP fundamentals, Benzinga news, your own journal, and the Alphalog research scorecard. You get answers grounded in data, not a model’s memory.
- Live quotes, fundamentals, and news via MCP tools
- Stock Intelligence 100-point scorecard on demand
- Session replay of every agent action for auditability
MCP ready
Bring the agent to your IDE.
Alphalog exposes an MCP server so you can use the same agent in Claude Desktop, Cursor, or any MCP-compatible client. Keys are managed from Settings with rotation and scope.
- Per-user MCP API keys with rotation
- Scoped access: read-only vs. full-tool
- Elite: higher rate limits and concurrent sessions
Who it's for
Active traders
End-of-day review
"Summarize today, flag my worst decision, and tell me what to do differently tomorrow." Alphalog answers with your actual trades.
Researchers
Deep-dive on demand
"Compare NVDA and AMD on fundamentals, technicals, and sentiment over the last 90 days." The agent assembles it in seconds.
Power users
Agent in your IDE
Connect Alphalog’s MCP server to Claude Desktop or Cursor and query your trading data from wherever you work.
Frequently asked
What model powers Alphalog AI?
We use frontier models (Claude) with journal-aware context and MCP tool use. Model details are surfaced in Settings for transparency.
Does it train on my data?
No. Your journal, trades, and conversations are never used to train shared models. Per-user isolation is enforced at the API level.
How is usage tracked?
Token usage and tool calls are logged per user. You see live counters in Settings and get alerts as you approach your tier’s limit.
Can I use Alphalog AI from Claude Desktop or Cursor?
Yes. Generate an MCP key from Settings and paste it into any MCP-compatible client. Same agent, your IDE.
What’s in Free vs. Pro vs. Elite?
Free includes light chat (journal-aware, limited tokens). Pro unlocks full agent with research tools. Elite adds higher limits, concurrent sessions, and priority model access.
Talk to the agent that knows your book.
Start your 14-day Pro trial. No credit card required.