lexforge-planlisted

<!-- model-block:start --> ## Model When your project names no model for your runtime, run this work on the model your own provider is given here. What the project names replaces this table. | Provider | Model | |---|---| | anthropic | claude-opus-5 | | openai | gpt-5.6-sol | | google | gemini-3.1-pro-preview | | deepseek | deepseek-v4-pro | | z.ai | glm-4.7 | A provider outside the table names nothing, so work on the model at work. <!-- model-block:end --> <!-- queue-rule:start --> ## Queue rule Run `lexforge status --change <name> --tool <your runtime> --json` first; parse stdout as JSON. Before it: no template, no questions, no files. No change named? Run `lexforge status --json` and ask which. Find your `id` in `artifacts`. Its `status` decides: - `ready` — work. The only status that lets you continue. - `blocked` — name every `blockedBy` id and `lexforge instructions <first blockedBy> --change <name> --tool <your runtime>`. Stop. - `done` — show `resolvedOutputPath`, ask before rewriting. - `skipped` — say `.lexforge.yaml` skips it, name `nextStep`. Stop. Once the artifact is written, run the command named in `nextStep` from the last machine response yourself, saying which one it is, and carry on into the next artifact. A question the artifact's own rule makes you ask is still asked, and its answer still waited for: this rule governs the handover between artifacts, nothing inside one. Whether to stop is read from `isPlanningComplete`, a field only `lexforge sta