Features
Multi-agent orchestration with Sisyphus, Prometheus, Atlas, Hephaestus
Ultra Work mode (ulw) for auto-planning, deep research, parallel agents, self-correction
Team Mode with up to 8 parallel agents and tmux visualization
Independent verification of all agent outputs
Session continuity via boulder.json after interruptions
3-tier MCP support (Model Context Protocol)
Background task execution
Category-based model routing (8 capability categories)
Intent gating and codebase assessment
Dynamic agent assembly per task
Skill injection for specialized capabilities (playwright, git-master, frontend, team-mode)
Parallel explore agents before code changes
Evidence-bound QA with Metis and Momus plan reviewers
Autonomous strategic report generation (Sakana Marlin)
Multi-agent LLM orchestration matching frontier models (Fugu Ultra)
Recursive Self-Improvement (RSI) Lab for autonomous AI research
Conductor system for natural language agent orchestration
Japan-based data residency and export control compliance
Finance multi-agent proposal generation with SMBC Group
AI-powered information analysis with DEEP DIVE
CoffeeBench benchmark for LLM agents in multi-agent economic environments
Autonomous model orchestration avoiding export control risks
Research-stage speech-to-speech AI with KAME architecture
