resumelisted

# Resume Orchestration The user wants to resume an interrupted orchestration. Follow this protocol to check for interrupted work, display a summary, and either resume or archive. ## Resume Protocol 1. **Check for interrupted work**: Read `.orchestray/state/orchestration.md`. - If the file exists and its frontmatter `status` is `in_progress` or `interrupted`: proceed to step 2. - If the file does not exist: Report "No interrupted orchestration found. Use `/orchestray:run [task]` to start a new orchestration." - If the file exists but `status` is `completed`: Report "Last orchestration completed successfully. Use `/orchestray:run [task]` for a new task." 2. **Build resume summary**: Read all files in `.orchestray/state/tasks/` directory. Parse each task file's YAML frontmatter to extract `id`, `title`, `status`, `assigned_to`, `depends_on`, `completed_at`. Read `.orchestray/state/orchestration.md` frontmatter for the original `task` description, `started_at`, `total_tasks`, and `completed_tasks`. Optionally read `.orchestray/state/agents/` to get agent run details for completed tasks. 3. **Display resume summary** in this format: ``` ## Interrupted Orchestration **Task:** {task from orchestration.md} **Started:** {started_at from orchestration.md} **Progress:** {completed_tasks}/{total_tasks} tasks ### Completed - [x] {task title} ({assigned_to}) -- {brief result from task file body} - [x] {task title} ({ass