Home / LLM agents & research
kyky2347/ALTA
Multi-agent LLM trading research platform with auditable execution boundary
What it is
A local, research-only multi-agent trading platform where LLM Scout agents pursue market leads, independent reviewers challenge the evidence, and a strategy desk weighs instruments and costs before any decision. Every handoff leaves an inspectable record, and execution offers a Shadow simulation mode plus a broker API mode that requires a verified, explicitly authorized account. Suits researchers and developers who want agent judgment separated from scheduling, data contracts and execution, with an operator console for auditing what ran. The project is explicitly experimental research software—not investment advice or evidence of profitable alpha, and live-account execution acceptance is not established.
At a glance
Research onlyOur rating, based on popularity, maintenance and how ready it is for real use.
| Best for | Professional quants |
|---|---|
| Used for | Strategy research, AI agents |
| Markets | Multi-market |
| Stack | Python |
| Learning curve | Steep learning curve |
| Practical value | Medium practical value |
| Cost | Free and open source |
| Hardware | No GPU needed |
| Maintenance | Last commit 4 days ago |
GitHub stars, last 30 days
Daily snapshots since 2026-09-12 (up to 30 days): +780 over the period, now 1,098. Gaps mean no snapshot was taken that day.
Autonomous LLM Trading Asterism — a research-only multi-agent trading platform with evidence-first research, an auditable console, Shadow simulation and explicitly authorized broker execution.