petri-multi-provider-generation

What problem does it solve?

This Skill standardizes how to build Petri nets that generate and coordinate code using multiple LLM providers in parallel, so the workflow stays consistent and reliably routes requests to the intended provider/model.

Core Features & Use Cases

• Multi-provider handler I/O contract: Defines the exact payload/key conventions that workers must read and write, including per-provider review keys.
• Provider/model selection rules: Enforces routing via ClaudeCode`ClaudeQueryBg with a pinned Model option, including distinctions between claude/anthropic/openai/lmstudio usage.
• Correct parallel composition semantics: Requires AND-merge for peer review so all reviewer outputs are present before continuing, preventing hidden failures.
• Failure-safe LLM error handling: Mandates treating any "Error:" response as handler failure ($Failed) to avoid false successes.

Quick Start

Ask your system to generate a Petri net that runs parallel reviews across Claude (via Claude Code CLI) and ChatGPT (via paid API when enabled), then merge the results only after both outputs arrive.