feat(autoplan): carve the four review phases + tasks aggregator into sections

Phase bodies (CEO/Design/Eng/DX consensus flows) and the Implementation Tasks
aggregator load on demand; Design and DX stay separate sections because each
is independently conditional on scope. Skeleton 83.7KB -> 58.7KB (-30%
always-loaded); the 6 decision principles, classification, sequencing, and
explicit skip-condition dispatch stay always-loaded. The chain E2E's
phase-complete markers now live only in sections, so its assertions double as
section-read proof (behavioral: external).

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

test/codex-web-search-flag.test.ts

@@ -67,10 +67,16 @@ describe('deprecated codex web-search flag is gone (#2525)', () => {
     }
   });
 
-  test('rendered autoplan skill resolves the token at every inline site', () => {
-    const rendered = fs.readFileSync(path.join(ROOT, 'autoplan', 'SKILL.md'), 'utf-8');
-    const count = rendered.split(CODEX_WEB_SEARCH_FLAG).length - 1;
-    expect(count).toBeGreaterThanOrEqual(4);
-    expect(rendered).not.toContain('{{CODEX_WEB_SEARCH_FLAG}}');
+  test('rendered autoplan phase sections resolve the token at every inline site', () => {
+    // The four phase bodies (and their codex invocations) are carved into
+    // autoplan/sections/*-phase.md — each generated section must carry the
+    // live flag, never the unresolved token.
+    for (const file of ['ceo-phase.md', 'design-phase.md', 'eng-phase.md', 'dx-phase.md']) {
+      const rendered = fs.readFileSync(path.join(ROOT, 'autoplan', 'sections', file), 'utf-8');
+      expect(rendered, `${file} lost the web-search flag`).toContain(CODEX_WEB_SEARCH_FLAG);
+      expect(rendered).not.toContain('{{CODEX_WEB_SEARCH_FLAG}}');
+    }
+    const skeleton = fs.readFileSync(path.join(ROOT, 'autoplan', 'SKILL.md'), 'utf-8');
+    expect(skeleton).not.toContain('{{CODEX_WEB_SEARCH_FLAG}}');
   });
 });

test/run-in-background-guidance.test.ts

@@ -18,8 +18,10 @@ import * as path from 'path';
 const ROOT = path.resolve(import.meta.dir, '..');
 
 // review's specialist-dispatch guidance lives in its carved Review Army section
-// (Step 4.5 moved out of the skeleton), so the pin follows it there.
-const GENERATED_WITH_GUIDANCE = ['review/sections/review-army.md', 'autoplan/SKILL.md'];
+// (Step 4.5 moved out of the skeleton), so the pin follows it there. Same for
+// autoplan: the dual-voice dispatch (Phase 1 override rules) lives in its
+// carved CEO-phase section.
+const GENERATED_WITH_GUIDANCE = ['review/sections/review-army.md', 'autoplan/sections/ceo-phase.md'];
 
 // The inverted, post-2.1.198-inert phrasings. Checked across every generated
 // SKILL.md so the regression can't migrate to another skill unnoticed.

test/skill-e2e-autoplan-chain.test.ts

@@ -85,7 +85,9 @@ describeE2E('/autoplan chain ordering (periodic)', () => {
 
           const budgetMs = 900_000; // 15 min
           const start = Date.now();
-          // Phase markers in autoplan/SKILL.md (lines 1126, 1211, 1331, 1437):
+          // Phase markers live in autoplan's carved phase sections
+          // (autoplan/sections/{ceo,design,eng,dx}-phase.md — the skeleton
+          // STOP-Reads each one at its phase boundary):
           //   "**Phase 1 complete." / "**Phase 2 complete." / "**Phase 3 complete." / "**Phase 3.5 complete."
           const phasePattern = /\*\*Phase\s+(\d+(?:\.\d+)?)\s+complete\.?\*\*/g;
 

test/skill-e2e-autoplan-dual-voice.test.ts

@@ -51,6 +51,14 @@ describeIfSelected('Autoplan dual-voice E2E', ['autoplan-dual-voice'], () => {
       const dest = path.join(skillsBase, skill);
       fs.mkdirSync(dest, { recursive: true });
       fs.copyFileSync(path.join(ROOT, skill, 'SKILL.md'), path.join(dest, 'SKILL.md'));
+      // Carved skills (autoplan + the plan-* reviews) keep their phase/review
+      // bodies in on-demand sections/ that the skeleton STOP-Reads — mirror the
+      // real install (which links sections/ next to SKILL.md) so the registered
+      // skeleton's section reads resolve inside the fixture.
+      const sections = path.join(ROOT, skill, 'sections');
+      if (fs.existsSync(sections)) {
+        copyDirSync(sections, path.join(dest, 'sections'));
+      }
     }
 
     // Write a tiny plan file for /autoplan to review.
@@ -139,8 +147,9 @@ Add a new /greet skill that prints a welcome message.
       // Full Phase 1 COMPLETION (three parallel review subagents, each loading a
       // 25-35K-token skill) routinely exceeds 10 minutes on sonnet, so requiring
       // the "Phase 1 complete" banner would force a 20-minute test for no extra
-      // dual-voice signal. Accept EITHER the completion banner (autoplan/SKILL.md
-      // "PHASE 1 COMPLETE" mandatory output) OR structural evidence that the
+      // dual-voice signal. Accept EITHER the completion banner (the "PHASE 1
+      // COMPLETE" mandatory output in autoplan/sections/ceo-phase.md, which the
+      // skeleton STOP-Reads) OR structural evidence that the
       // Phase 1 review dispatch actually happened: an Agent tool_use whose input
       // carries review instructions (execution artifact built by the skill, not
       // an echo of our prompt).