grade-iterate

原文涵盖的主题

• Phase 3 — Grade → Iterate (the bounded loop)
• Workflow
• Hard rules
• Forcing-question library (recommend + cite)
• Tools

This is the plugin's loop: CMA's outcome primitive self-grades the agent's work in an isolated context and feeds failing verdicts back for the next attempt. It is always bounded by maxiterations (1..20) — never "improve forever". See ../../references/loops-and-workflows.md and the outcome section of ../../references/cma-primitives.md. 1. Define the outcome. The rubric is required; maxiterations is clamped to 1..20. Send the payload as a user.defineoutcome event (append to the running session). 2. Read every verdict first. Tables the rubric outcome and recommends: SHIP (satisfied), SHARPEN then re-run (needsrevision), ESCALATE (maxiterationsreached / failed), RESUME (interrupted). With ≤1 iteration left it flips to "make the single highest-value fix or escalate now". 3. Loop invariant. Each iteration must move ≥1 rubric line fail→pass, or the run halts at the cap and escalates. Don't burn the budget on cosmetic edits. 4. Once a version passes, run held-back eval. Held-back cases (never seen during iteration) run in parallel, capped at the 25-thread CMA ceiling, each graded against the same rubric. 5. Decide. SHIP as v0, or promote to a scheduled deployment (Phase 4). Record the verd

这里只展示节选，请以完整原文和当前要求为准。