rote

原文涵盖的主题

• rote: compile a skill into a deterministic pipeline
• When this applies
• 1. Identify the source skill
• 2. Pick a runtime target
• 3. Resolve the CLI
• 4. Run the compilation
• 5. Report the result
• 6. Serve compiled pipelines back to Claude

You orchestrate the rote CLI. It runs an LLM compiler agent over a source skill once, and emits a pipeline that runs forever after without an agent loop. Your job is to resolve the inputs, run the CLI, and interpret the output. You never classify nodes or write pipeline.yaml yourself; the CLI's compiler agent does that. Use it on a skill the user has already run many times and wants to run many more, unattended. Exploratory or one-off work should stay an agent loop: flexibility is the point there, and there is nothing proven to compile yet. Say so and stop if that is what you are looking at. The source is a directory containing a SKILL.md, optionally with a references/ folder. The user names it, or you infer it from context: a skill just discussed, a path in the conversation, .claude/skills/ or skills/ in the project. Confirm the resolved absolute path with the user before running. Compilation costs real time and tokens, so never guess and go. If the directory has no SKILL.md, stop and ask. | Runtime | --runtime | Language | Choose when | | --- | --- | --- | --- | | DBOS (default) | dbos | Python | No orchestrator to deploy. SQLite for dev, Postgres for prod | | Temporal | temporal

这里只展示节选，请以完整原文和当前要求为准。