exec-dispatch

# exec-dispatch ## Overview Execute an approved plan one task at a time. Each task runs in a **fresh subagent** with no prior conversation history — only the RTCO brief and the plan excerpt. On return, the output passes through schema validation, self-review, and plan-compliance review before the task is marked complete. One task per agent. One commit per task. No batching. ## When to invoke After `plan-write` has saved a plan and the user has chosen the subagent-driven execution option. Do not invoke if: - No plan file exists. - The user chose inline execution (a different skill handles that). - Tasks in the plan are incomplete or contain placeholders — stop, surface the issue. ## Prerequisites - Plan at `docs/plans/YYYY-MM-DD-<feature-name>.md`, approved. - Work happens in a dedicated worktree or branch. - The repo is in a clean state (no uncommitted work that isn't part of this run). If any prerequisite fails, stop and report what is missing. ## Dispatch loop For each task in the plan, in order: 1. **Select the task.** Read the plan file. Take the next unchecked task. Copy its full content (files list + all steps) verbatim. Do not paraphrase. 1a. **Verify-clause gate.** The selected task header MUST end with `→ verify: <criterion>`. If absent, vague ("works", "looks right"), or non-checkable, **REJECT the task**: stop the loop, surface the gap to the user, and route back to `plan-write` to repair the plan. Do not dispatch tasks lacking a verify clause. 2. **Com...