#7575th of 89 in Agent platformsFor finance and business, trading, stocks and crypto, finance and accounting
alta
Multi-agent LLM market research system with simulated and broker execution
- Stars
- 1.1k
- License
- Apache-2.0
- Last commit
- Sep 2026
- Last release
- Aug 2026
- Language
- Python
0 on 25 Aug 20261,077 on 10 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
ALTA runs four specialist LLM Scouts that research market opportunities. Each thesis then goes to independent reviewers, code-based ranking and a strategy desk that compares instruments. Every handoff is saved as a versioned record in PostgreSQL and shown in a React operator console. Execution is either an internal Shadow simulation or a broker API (Tiger, Alpaca, IBKR, Futu). The README labels it research-only software.
Who it is for: Researchers and developers studying multi-agent financial research workflows
Strengths
- Reviewers commit assessments independently; code, not an Agent, handles ranking and execution gates
- Evidence, reviews and outcomes are linked records traceable by ID, with event time kept separate
- Shadow mode simulates fills and exits without sending broker orders
- Console in English, Simplified and Traditional Chinese; model assignable per agent role
Weaknesses
- Live-account acceptance not established; Longbridge and Schwab connectors are authorization-blocked
- Broker lifecycle limited to one plan per account, long USD stocks/ETFs, whole shares, DAY limits
- Exits are software-managed, so downtime can delay them; stopping ALTA leaves positions open
- Needs Node 22, Python 3.12, uv, Docker and a Rust toolchain; macOS or Linux only
What it needs
- Needs PostgreSQL, Redis, Docker/OrbStack, Node.js 22+, Python 3.12, uv, Rust toolchain, Codex App Server
- Models: Codex App Server sessions, configurable models per agent role
What changed
From the release notes. No flag means the note does not say so, not that the upgrade is safe.
Also in Agent platforms
See all 89| Rank | Project | Score |
|---|---|---|
| 1 | AutoGPTRole-based AI agents coordinated by one lead agent, plus a visual workflow builder | 95 out of 100 |
| 2 | daily_stock_analysisLLM-driven daily analysis of watchlist stocks across six markets, pushed to chat apps | 95 out of 100 |
| 3 | LangflowVisual builder for AI agents and workflows, exposed as API or MCP server | 94 out of 100 |
| #4 | deer-flowAgent harness running sub-agents, memory and sandboxes with pluggable skills | 93 out of 100 |
| #5 | DifyVisual builder for LLM workflows, RAG pipelines and agents, with APIs | 92 out of 100 |