Pinned Loading

1. crucible crucible Public

   AI-native multi-agent research workflow: parallel evidence gathering, 7-direction debate, and risk-gated analysis — structured output, not one-shot prompts.

   Python 108 11

2. AlphaAirlock AlphaAirlock Public

   Agentic multi-agent quant research system: LLM agents generate, sandbox-backtest, paper-trade, and gate alpha strategies before any live routing (live off by default). FastAPI + Next.js, multi-asse…

   Python

3. stage0-agent-runtime-guard stage0-agent-runtime-guard Public

   Customer-facing demo showing why AI agents need runtime policy authority before publishing, deploying, or creating side effects.

   Python 1

4. stage0-execution-guard-skill stage0-execution-guard-skill Public

   Fail-closed execution gatekeeper for AI agents. Enforces Stage0 authorization before any action.

   Python 1

5. itechmeat/open-second-brain itechmeat/open-second-brain Public

   Local-first 🧠 memory for Hermes Agent that lives in your Obsidian vault and remembers project context. Nightly 😴 dream passes turn repeat corrections into confirmed preferences with measurable conf…

   TypeScript 125 7

6. polakowo/vectorbt polakowo/vectorbt Public

   The backtesting engine that gives you an unfair advantage. Run thousands of trading ideas before others finish one.

   Python 8.3k 1.1k