test: clean up the paid eval lane (B1-B4, B6, B7)

- B1: delete paid files that assert nothing or cannot pass meaningfully:
  skill-llm-eval-spec and skill-e2e-spec-execute (test.todo), gemini-e2e
  (+ gemini-session-runner; no gemini CLI in CI), ship-idempotency (red
  since v1.63), the two opus-4-7 *-sonnet overlay wrappers, conductor-prose
  (+ its source-evaluation replay), codex-e2e-plan-format; drop their keys,
  scripts and census rows.
- B2: skill-llm-eval grades browse/sections/command-list.md with one union
  judge that also carries the baseline score pin; regression-vs-baseline
  deleted (paid run: pass, c4/c4/a4).
- B3: memory-pipeline, ios-qa, ios-qa-swift-build and plan-tune-cathedral
  make no model calls; renamed out of the paid glob so they run on every
  PR. Swift builds need GSTACK_TEST_SWIFT=1; device stub deleted.
- B4: codex-e2e*, outside-voice, aside and ios-device cannot run in the CI
  image; excluded from the weekly lane with a tracked re-entry condition.
- B6: fold opus-47's negative routing controls into skill-routing-e2e
  journey-negatives (paid run: 3/3 unrouted) and delete the file.
- B7: delete the never-green brain-privacy-gate eval; a free
  gstack-skill-start test now proves consent precedes artifacts egress.

test/skill-llm-eval.test.ts

@@ -12,7 +12,6 @@
 
 import { afterAll, expect } from 'bun:test';
 import { JUDGE_MS } from './helpers/eval-budgets';
-import Anthropic from '@anthropic-ai/sdk';
 import * as fs from 'fs';
 import * as path from 'path';
 import { callJudge, judge, JudgeRefusalError, DEFAULT_JUDGE_MAX_TOKENS } from './helpers/llm-judge';
@@ -62,12 +61,11 @@ function readBrowseCommandSection(): string {
 }
 
 /** Slice a section out of the command-list section file, guarded non-empty. */
-function sliceBrowseSection(startHeader: string, endHeader?: string): string {
+function sliceBrowseSection(startHeader: string): string {
   const content = readBrowseCommandSection();
   const start = content.indexOf(startHeader);
   if (start < 0) throw new Error(`browse/sections/command-list.md: "${startHeader}" not found`);
-  const end = endHeader ? content.indexOf(endHeader) : -1;
-  const section = end > start ? content.slice(start, end) : content.slice(start);
+  const section = content.slice(start);
   if (section.trim().length < 200) {
     throw new Error(`browse/sections/command-list.md slice at "${startHeader}" is empty/stub — regenerate with: bun run gen:skill-docs`);
   }
@@ -91,85 +89,45 @@ function testIfSelected(testName: string, fn: () => Promise<void>, timeout: numb
 }
 
 describeIfSelected('LLM-as-judge quality evals', [
-  'command reference table', 'snapshot flags reference',
-  'browse/SKILL.md reference', 'setup block', 'regression vs baseline',
+  'browse/SKILL.md reference', 'setup block',
 ], () => {
-  testIfSelected('command reference table', async () => {
-    const t0 = Date.now();
-    // Browse carve: the command reference lives in the generated on-demand
-    // section browse/sections/command-list.md now (read via non-empty guard).
-    const section = sliceBrowseSection('## Full Command List');
-
-    const scores = await judge('command reference table', section);
-    console.log('Command reference scores:', JSON.stringify(scores, null, 2));
-
-    // Completeness threshold is 3 (not 4) — the command reference table is
-    // intentionally terse (quick-reference format). The judge consistently scores
-    // completeness=3 because detailed argument docs live in per-command sections.
-    evalCollector?.addTest({
-      name: 'command reference table',
-      suite: 'LLM-as-judge quality evals',
-      tier: 'llm-judge',
-      passed: scores.clarity >= 3 && scores.completeness >= 3 && scores.actionability >= 4,
-      duration_ms: Date.now() - t0,
-      cost_usd: 0.02,
-      judge_scores: { clarity: scores.clarity, completeness: scores.completeness, actionability: scores.actionability },
-      judge_reasoning: scores.reasoning,
-    });
-
-    expect(scores.clarity).toBeGreaterThanOrEqual(3);
-    expect(scores.completeness).toBeGreaterThanOrEqual(3);
-    expect(scores.actionability).toBeGreaterThanOrEqual(4);
-  }, JUDGE_MS);
-
-  testIfSelected('snapshot flags reference', async () => {
-    const t0 = Date.now();
-    // Browse carve: snapshot flags live in browse/sections/command-list.md now,
-    // ordered before '## Full Command List' (the '## CSS Inspector' end boundary
-    // stayed in the skeleton).
-    const section = sliceBrowseSection('## Snapshot Flags', '## Full Command List');
-
-    const scores = await judge('snapshot flags reference', section);
-    console.log('Snapshot flags scores:', JSON.stringify(scores, null, 2));
-
-    evalCollector?.addTest({
-      name: 'snapshot flags reference',
-      suite: 'LLM-as-judge quality evals',
-      tier: 'llm-judge',
-      passed: scores.clarity >= 3 && scores.completeness >= 4 && scores.actionability >= 4,
-      duration_ms: Date.now() - t0,
-      cost_usd: 0.02,
-      judge_scores: { clarity: scores.clarity, completeness: scores.completeness, actionability: scores.actionability },
-      judge_reasoning: scores.reasoning,
-    });
-
-    expect(scores.clarity).toBeGreaterThanOrEqual(3);
-    expect(scores.completeness).toBeGreaterThanOrEqual(4);
-    expect(scores.actionability).toBeGreaterThanOrEqual(4);
-  }, JUDGE_MS);
-
   testIfSelected('browse/SKILL.md reference', async () => {
     const t0 = Date.now();
-    // Browse carve: flags + commands are the whole generated section file.
+    // Browse carve: snapshot flags + the full command list are the whole
+    // generated section file; one judge grades the union. Scores are also
+    // pinned against test/fixtures/eval-baselines.json (UPDATE_BASELINES=1
+    // rewrites the pin).
     const section = sliceBrowseSection('## Snapshot Flags');
 
     const scores = await judge('browse skill reference (flags + commands)', section);
     console.log('Browse SKILL.md scores:', JSON.stringify(scores, null, 2));
 
+    const baselinesPath = path.join(ROOT, 'test', 'fixtures', 'eval-baselines.json');
+    const baselines = JSON.parse(fs.readFileSync(baselinesPath, 'utf-8'));
+    const regressions = (['clarity', 'completeness', 'actionability'] as const)
+      .filter(dim => scores[dim] < baselines.browse_skill[dim])
+      .map(dim => `browse_skill.${dim}: ${scores[dim]} < baseline ${baselines.browse_skill[dim]}`);
+    if (process.env.UPDATE_BASELINES) {
+      baselines.browse_skill = { clarity: scores.clarity, completeness: scores.completeness, actionability: scores.actionability };
+      fs.writeFileSync(baselinesPath, JSON.stringify(baselines, null, 2) + '\n');
+      console.log('Updated eval baselines');
+    }
+
     evalCollector?.addTest({
       name: 'browse/SKILL.md reference',
       suite: 'LLM-as-judge quality evals',
       tier: 'llm-judge',
-      passed: scores.clarity >= 3 && scores.completeness >= 4 && scores.actionability >= 4,
+      passed: scores.clarity >= 3 && scores.completeness >= 4 && scores.actionability >= 4 && regressions.length === 0,
       duration_ms: Date.now() - t0,
       cost_usd: 0.02,
       judge_scores: { clarity: scores.clarity, completeness: scores.completeness, actionability: scores.actionability },
-      judge_reasoning: scores.reasoning,
+      judge_reasoning: regressions.length ? `${scores.reasoning} | ${regressions.join('; ')}` : scores.reasoning,
     });
 
     expect(scores.clarity).toBeGreaterThanOrEqual(3);
     expect(scores.completeness).toBeGreaterThanOrEqual(4);
     expect(scores.actionability).toBeGreaterThanOrEqual(4);
+    expect(regressions).toEqual([]);
   }, JUDGE_MS);
 
   testIfSelected('setup block', async () => {
@@ -205,89 +163,6 @@ describeIfSelected('LLM-as-judge quality evals', [
     expect(scores.clarity).toBeGreaterThanOrEqual(3);
   }, JUDGE_MS);
 
-  testIfSelected('regression vs baseline', async () => {
-    const t0 = Date.now();
-    // Browse carve: the command reference lives in browse/sections/command-list.md.
-    const genSection = sliceBrowseSection('## Full Command List');
-
-    const baseline = `## Command Reference
-
-### Navigation
-| Command | Description |
-|---------|-------------|
-| \`goto <url>\` | Navigate to URL |
-| \`back\` / \`forward\` | History navigation |
-| \`reload\` | Reload page |
-| \`url\` | Print current URL |
-
-### Interaction
-| Command | Description |
-|---------|-------------|
-| \`click <sel>\` | Click element |
-| \`fill <sel> <val>\` | Fill input |
-| \`select <sel> <val>\` | Select dropdown |
-| \`hover <sel>\` | Hover element |
-| \`type <text>\` | Type into focused element |
-| \`press <key>\` | Press key (Enter, Tab, Escape) |
-| \`scroll [sel]\` | Scroll element into view |
-| \`wait <sel>\` | Wait for element (max 10s) |
-| \`wait --networkidle\` | Wait for network to be idle |
-| \`wait --load\` | Wait for page load event |
-
-### Inspection
-| Command | Description |
-|---------|-------------|
-| \`js <expr>\` | Run JavaScript |
-| \`css <sel> <prop>\` | Computed CSS |
-| \`attrs <sel>\` | Element attributes |
-| \`is <prop> <sel>\` | State check (visible/hidden/enabled/disabled/checked/editable/focused) |
-| \`console [--clear\\|--errors]\` | Console messages (--errors filters to error/warning) |`;
-
-    const client = new Anthropic();
-    const response = await client.messages.create({
-      model: 'claude-sonnet-4-6',
-      max_tokens: 1024,
-      messages: [{
-        role: 'user',
-        content: `You are comparing two versions of CLI documentation for an AI coding agent.
-
-VERSION A (baseline — hand-maintained):
-${baseline}
-
-VERSION B (auto-generated from source):
-${genSection}
-
-Which version is better for an AI agent trying to use these commands? Consider:
-- Completeness (more commands documented? all args shown?)
-- Clarity (descriptions helpful?)
-- Coverage (missing commands in either version?)
-
-Respond with ONLY valid JSON:
-{"winner": "A" or "B" or "tie", "reasoning": "brief explanation", "a_score": N, "b_score": N}
-
-Scores are 1-5 overall quality.`,
-      }],
-    });
-
-    const text = response.content[0].type === 'text' ? response.content[0].text : '';
-    const jsonMatch = text.match(/\{[\s\S]*\}/);
-    if (!jsonMatch) throw new Error(`Judge returned non-JSON: ${text.slice(0, 200)}`);
-    const result = JSON.parse(jsonMatch[0]);
-    console.log('Regression comparison:', JSON.stringify(result, null, 2));
-
-    evalCollector?.addTest({
-      name: 'regression vs baseline',
-      suite: 'LLM-as-judge quality evals',
-      tier: 'llm-judge',
-      passed: result.b_score >= result.a_score,
-      duration_ms: Date.now() - t0,
-      cost_usd: 0.02,
-      judge_scores: { a_score: result.a_score, b_score: result.b_score },
-      judge_reasoning: result.reasoning,
-    });
-
-    expect(result.b_score).toBeGreaterThanOrEqual(result.a_score);
-  }, JUDGE_MS);
 });
 
 // --- Part 7: QA skill quality evals (C6) ---
@@ -523,59 +398,6 @@ score (1-5): 5 = perfectly consistent, 1 = contradictory`);
   }, JUDGE_MS);
 });
 
-// --- Part 7: Baseline score pinning (C9) ---
-
-describeIfSelected('Baseline score pinning', ['baseline score pinning'], () => {
-  const baselinesPath = path.join(ROOT, 'test', 'fixtures', 'eval-baselines.json');
-
-  testIfSelected('baseline score pinning', async () => {
-    const t0 = Date.now();
-    if (!fs.existsSync(baselinesPath)) {
-      console.log('No baseline file found — skipping pinning check');
-      return;
-    }
-
-    const baselines = JSON.parse(fs.readFileSync(baselinesPath, 'utf-8'));
-    const regressions: string[] = [];
-
-    // Browse carve: the command reference lives in browse/sections/command-list.md.
-    const cmdSection = sliceBrowseSection('## Full Command List');
-    const cmdScores = await judge('command reference table', cmdSection);
-
-    for (const dim of ['clarity', 'completeness', 'actionability'] as const) {
-      if (cmdScores[dim] < baselines.command_reference[dim]) {
-        regressions.push(`command_reference.${dim}: ${cmdScores[dim]} < baseline ${baselines.command_reference[dim]}`);
-      }
-    }
-
-    if (process.env.UPDATE_BASELINES) {
-      baselines.command_reference = {
-        clarity: cmdScores.clarity,
-        completeness: cmdScores.completeness,
-        actionability: cmdScores.actionability,
-      };
-      fs.writeFileSync(baselinesPath, JSON.stringify(baselines, null, 2) + '\n');
-      console.log('Updated eval baselines');
-    }
-
-    const passed = regressions.length === 0;
-    evalCollector?.addTest({
-      name: 'baseline score pinning',
-      suite: 'Baseline score pinning',
-      tier: 'llm-judge',
-      passed,
-      duration_ms: Date.now() - t0,
-      cost_usd: 0.02,
-      judge_scores: { clarity: cmdScores.clarity, completeness: cmdScores.completeness, actionability: cmdScores.actionability },
-      judge_reasoning: passed ? 'All scores at or above baseline' : regressions.join('; '),
-    });
-
-    if (!passed) {
-      throw new Error(`Score regressions detected:\n${regressions.join('\n')}`);
-    }
-  }, JUDGE_MS);
-});
-
 // --- Workflow SKILL.md quality evals (10 new tests for 100% coverage) ---
 
 /**