docs(voyage): S20-fix — correct stale "Use Sonnet for all sub-agents" in 3 orchestrator docs

Follow-up to e7a58b0 (S20). Operator directive: errors found during a
task must always be fixed, not flagged out-of-scope. The three
orchestrator reference docs (review/research/planning) carried a
pre-pinning "Cost: Use Sonnet for all sub-agents" rule that contradicts
the operator-pinned all-opus reality (40d8742 — all 24 named agents are
model: opus, guarded by agent-frontmatter.test.mjs). The authoritative
command (trekresearch.md) was already correct.

Corrected all three to mirror trekresearch.md: sub-agents use their
pinned model: frontmatter (currently opus); a phase_signals model signal
or the active --profile (e.g. economy) overrides per-phase. Legitimate
sonnet uses (economy profile, low-effort path) are per-phase overrides,
not a blanket default — left untouched. The two "(Sonnet, Explore)"
ad-hoc follow-up/deep-dive hints are outside the 24-agent pinning and
align with the global "Sonnet for retrieval" model strategy — left as-is.

TDD: anti-false-claim doc-consistency pin written failing-first (3 hits)
in doc-consistency.test.mjs, then the prose fixed. Suite 713->714
(712 pass / 2 skip / 0 fail); census doc-pins 69->70, total 670->671.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LqBYc8Ltrk7LipyJmGxXiB

agents/planning-orchestrator.md

@@ -477,7 +477,10 @@ You can:
   of the brief (Intent, Goal, Constraint, Preference, NFR, Success Criterion).
   A plan step with no brief basis is scope creep — flag it or remove it.
 - **Scope:** Only explore the current working directory. Never read files outside the repo.
-- **Cost:** Use Sonnet for all sub-agents. You (the orchestrator) run on Opus.
+- **Cost:** Sub-agents use their pinned `model:` frontmatter (currently `opus`
+  for all). A `phase_signals[<phase>].model` brief signal or the active
+  `--profile` (e.g. `economy`) overrides the model per-phase. You (the
+  orchestrator) run on Opus.
 - **Privacy:** Never log secrets, tokens, or credentials.
 - **Quality:** Every file path in the plan must be verified. Every "reuses" reference
   must point to real code. The plan must stand alone without exploration context.

agents/research-orchestrator.md

@@ -218,7 +218,10 @@ You can:
 
 - **Scope:** Codebase analysis is limited to the current working directory.
   External research has no such limit.
-- **Cost:** Use Sonnet for all sub-agents. You (the orchestrator) run on Opus.
+- **Cost:** Sub-agents use their pinned `model:` frontmatter (currently `opus`
+  for all). A `phase_signals[<phase>].model` brief signal or the active
+  `--profile` (e.g. `economy`) overrides the model per-phase. You (the
+  orchestrator) run on Opus.
 - **Privacy:** Never log secrets, tokens, or credentials in the brief.
 - **Sources:** Every claim in the brief must cite a source (URL or file path).
   Never invent findings.

agents/review-orchestrator.md

@@ -242,8 +242,10 @@ Append a stats line to `${CLAUDE_PLUGIN_DATA}/trekreview-stats.jsonl`:
 - **No silent drops.** Every file in the discovered diff must appear in
   the Coverage section, even if its treatment is `skip`. Hidden truncation
   is COVERAGE_SILENT_SKIP (MAJOR).
-- **Cost:** Use Sonnet for all sub-agents. The orchestrator (the
+- **Cost:** Sub-agents use their pinned `model:` frontmatter (currently `opus`
-  /trekreview command itself) runs on Opus.
+  for all). A `phase_signals[<phase>].model` brief signal or the active
+  `--profile` (e.g. `economy`) overrides the model per-phase. The orchestrator
+  (the /trekreview command itself) runs on Opus.
 - **Privacy:** Never log secrets, tokens, or credentials. Findings citing
   files with secret-like content must redact the secret in the `detail`.
 - **Honesty:** If the diff is trivially small or all-skip, say so. Do

tests/lib/doc-consistency.test.mjs

@@ -304,6 +304,25 @@ test('Phase 9 prose mandates parallel single-message dispatch + inline dedup', (
   );
 });
 
+// --- S20 — anti-false-claim: orchestrator docs must not claim ALL sub-agents run on Sonnet ---
+//
+// All named swarm agents are pinned `model: opus` (operator decision 40d8742;
+// structural source-of-truth = agent-frontmatter.test.mjs). The orchestrator
+// reference docs carried a pre-pinning "Use Sonnet for all sub-agents" cost rule
+// that contradicts that pinning. This anti-false-claim pin keeps the prose honest.
+// Legitimate sonnet uses (economy profile; low-effort `phase_signals.model`) are
+// per-phase overrides, NOT a blanket "all sub-agents" default — unaffected here.
+test('orchestrator docs do not claim ALL sub-agents run on Sonnet (agents are model: opus) [S20]', () => {
+  const orchestrators = listMd('agents').filter(f => f.endsWith('orchestrator.md'));
+  assert.ok(orchestrators.length >= 3, `expected ≥ 3 orchestrator reference docs, got ${orchestrators.length}`);
+  for (const f of orchestrators) {
+    assert.ok(
+      !/Use Sonnet for all sub-agents/i.test(read(`agents/${f}`)),
+      `agents/${f}: stale "Use Sonnet for all sub-agents" contradicts the all-opus agent pinning — fix the prose, not this test`,
+    );
+  }
+});
+
 // --- v4.1 Step 21 — pin --profile + phase_models on the 6 commands ---
 //
 // CLAUDE.md / README.md pinning is deferred to Step 22 (post-write of