Back to all projects
AI/ML2025

Agents Platform

Local-first multi-agent orchestration with human-gated risk tiers

TypeScriptNode.jsLangGraphOpenAIAnthropic

Problem

Multi-agent systems need proper orchestration, risk management, and human oversight. Existing frameworks lack fine-grained control over agent actions and risk tiers.

Approach

Built a platform with structured task-thread collaboration, human-gated risk tiers, and adapter model for multiple LLM runtimes. Local-first with snapshot persistence for restart recovery.

Result

Multi-agent platform with OpenAI, Anthropic, and Ollama runtimes. Human-gated risk tiers and local snapshot persistence.