What problem does it solve? Coordinating multiple AI agents for complex tasks usually requires external orchestration frameworks and custom code. This Skill turns the AI session itself into an OpenProse virtual machine that runs structured .prose programs, spawning subagents, managing state, and handling control flow without external dependencies. ## Core Features & Use Cases - Program Execution: Run local .prose files or fetch programs from URLs and the p.prose.md registry, with the AI embodying the VM and spawning real subagent sessions. - Compilation & Validation: Validate .prose syntax against the full language grammar, checking agent definitions, sessions, loops, pipelines, and error handling constructs. - State Management: Choose between filesystem, in-context, SQLite, or PostgreSQL state backends, with persistent agents that retain memory across runs. - Use Case: Write a .prose program that runs three code reviews in parallel, synthesizes the findings, and loops until the code passes quality checks, then execute it with a single prose run command. ## Quick Start Run the hello world example by asking the AI to execute prose run examples/01-hello-world.prose.
