Orchestrate and execute OpenProse programs for multi-agent workflows.
npx skills add https://github.com/Yogi-076/Vajrascan.on --skill prose-yogi-076
Please help me install this Agent Skill. Skill: prose Source: https://github.com/Yogi-076/Vajrascan.on/tree/main/moltbot/extensions/open-prose/skills/prose Command: npx skills add https://github.com/Yogi-076/Vajrascan.on --skill prose-yogi-076
OpenProse provides a VM and language for orchestrating multi-agent AI workflows so users can author, run, and manage complex .prose programs without manually coordinating subagents, state, or bindings. It removes the friction of spawning many model sessions, persisting intermediate results, and composing modular programs from examples or remote registry entries.
Invoke the OpenProse VM by running prose run examples/01-hello-world.prose to execute the bundled hello-world example.
High-intent search queries and answers about installing and using this skill.
Multi-agent orchestration is handled by executing OpenProse programs, which spawn Task tool sessions and narrate execution state automatically. This removes the friction of manually coordinating subagents, persisting intermediate results, or composing modular programs.
Persistent agent memory supports filesystem, in-context, SQLite, and PostgreSQL state backends. These backends manage bindings, scoped execution IDs, and persistent memory to maintain state across multi-agent workflow sessions.
Remote .prose programs run by resolving handle/slug shorthand and URL sources through registry-based remote fetch resolution. This allows you to import and execute local or registry-hosted programs modularly without manual downloads.
Recursive language-model pipelines support parallel blocks, loops, retries with backoff, choice/conditionals, and recursion for large-scale orchestration. These control flow mechanisms enable complex RLM-style patterns within the OpenProse VM.
Multi-agent workflow state management does not require a specific database setup. You can choose between filesystem, in-context, SQLite, or PostgreSQL backends depending on your persistence and scalability requirements.
To start running a basic multi-agent workflow, invoke the OpenProse VM by executing a command like `prose run examples/01-hello-world.prose`. This runs the bundled hello-world example to validate your setup and demonstrate execution semantics.
MaxSaietsExecute .prose programs with a VM-like runtime for multi-agent workflows.
letta-aiCompile SQLite with gcov instrumentation and verify .gcno/.gcda generation.
gmotylControl Claude Code sessions remotely via Telegram with persistent bindings.
gensparxDefine and execute multi-agent workflows using the OpenProse programming language.
