Auditable by design
Stable markdown reports with deterministic section headings, structured metadata, and a daily index. Built for human review and downstream machine ingestion alike.
Agent-Alpha is a LangGraph-powered multi-agent pipeline that turns one or many ticker symbols into deterministic markdown and JSON research reports. Specialist analysts, adversarial researchers, a trader, a risk committee, and a portfolio manager collaborate through structured disagreement to produce auditable, five-tier investment recommendations.
Most trading systems are rigid rule engines or opaque ML models. Agent-Alpha sits in the middle: every recommendation comes with an inspectable reasoning trail, from analyst findings through adversarial debate to final decision.
Stable markdown reports with deterministic section headings, structured metadata, and a daily index. Built for human review and downstream machine ingestion alike.
The system intentionally manufactures disagreement. Bull and bear researchers debate, then three risk debaters pressure-test exposure before the portfolio manager decides.
An append-only decision log tracks outcomes, computes alpha versus SPY, and reflects lessons back into future runs. Lightweight learning without fine-tuning new weights.
A directed LangGraph workflow stages analysts, debate, synthesis, risk review, and a final decision into a single deterministic pass.
Four specialists (market, sentiment, news, fundamentals) collect tool-backed evidence and write independent reports.
Each analyst cycles through LangChain tool nodes until continuation logic is satisfied, then messages are cleared.
Two structurally opposed researchers debate across configurable rounds, sharpening disagreement into synthesis.
Research manager writes the plan, trader proposes action, and three risk debaters cycle aggressive/conservative/neutral.
The PM receives the full chain plus memory-log context, writes the final rating, and logs state for future reflection.
Named roles, not anonymous prompt fragments. Each agent occupies a specific graph node, uses the quick or deep model profile, and contributes a distinct artifact.
Price action, indicators, and technical trend context.
Public mood and positioning signals from news-backed sentiment.
Company and macro developments, global news, insider activity.
Financial statements, cash flow, income statement, balance sheet.
Constructs and defends the strongest upside case from analyst reports.
Challenges optimism and argues the downside, forcing robust synthesis.
Synthesizes debate into a structured plan with five-tier rating scale.
Turns the research plan into an actionable proposal with position guidance.
Pushes for conviction and higher exposure where thesis is strong.
Focuses on downside, fragility, and event-risk containment.
Balances aggressive and defensive positions to keep committee grounded.
Receives the full chain plus memory-log context and issues the final rating.
The repo is organized around a script-first operational core, a minimal local web app, structured planning docs, and a comprehensive test suite.
Installable Python package with graph orchestration, agent roles, LLM provider adapters, dataflow helpers, deterministic reporting, and shared config.
The primary interface for running analyses. Supports single and batch tickers, provider selection, model override, debate rounds, and checkpoint resume.
Lightweight stdlib HTTP server and static frontend for launching runs, saving defaults and API keys, monitoring jobs, and browsing generated reports.
Product overview, output schema contract, and a SaaS Studio vision that maps the evolution from a local script wrapper to a production AI operations platform.
Reports follow a strict schema contract designed for both human review and machine ingestion. Every field has a defined type, location, and validation level.
report.mdCanonical analysis report with 11 required sections.
metadata.jsonMachine-parseable run metadata mirroring report primitives.
Date-first folder structure with daily index aggregation.
The planning docs describe the evolution from script-first local execution to a polished SaaS AI operations platform with live runs, agent discovery, report history, and enterprise controls.