chore(skills): trim sdlc/update SKILL.md below 5K-token threshold (v1.48.0)

[BaseInfinity]

Summary

PR #272's scripts/audit-session-load.sh flagged 2 of 4 SKILL.md files as TRIM candidates the same day we shipped the tool:

• skills/sdlc/SKILL.md: 12,427 tokens
• skills/update/SKILL.md: 8,555 tokens

Acted on the tool's findings — closes the Prove-It loop. A tool that surfaces real issues whose owner ignores them is just a louder lint warning.

Result:

• skills/sdlc/SKILL.md: 12,427 → 4,995 tokens (-60%, 49,709 → 19,983 chars)
• skills/update/SKILL.md: 8,555 → 4,044 tokens (-53%, 34,220 → 16,179 chars)
• Audit reports 0 trim candidates (was 2)

What was trimmed

No operational content lost. Aggressive prose compression — removed ASCII-art decoration boxes (kept the bold sentences they contained), tightened cross-model review section while preserving every Codex command/sandbox note/dialogue-loop template/convergence rule. TodoWrite checklist intact (all 30 items, with activeForm removed since the spinner falls back to subject). Step 1.5's 30-line Node cmp() helper replaced with a precise prose description of the algorithm. Step 3's frozen 20-line changelog example shortened to a placeholder pointing at the real fetched CHANGELOG.

Every test anchor traced and preserved: all grep'd phrases across tests/test-{doc-consistency,self-update,update-skill-step-7-7,update-skill-cli-version,memory-audit-protocol,docs-usability,cli,prove-it,hooks}.sh still match.

New quality test

test_wizard_own_skills_below_threshold in tests/test-audit-session-load.sh — runs the audit on the wizard repo itself and fails if any SKILL.md flags TRIM. RED before this PR (both files flagged), GREEN after. Mutation-verifiable: bumping either file ~200 tokens flips the test red.

Codex review

Round 1 CERTIFIED 10/10. No findings. Codex did its own RED/GREEN proof (stashed only the trimmed skill files to keep the new test active), verified every checklist item with shell evidence, ran the full CONTRIBUTING.md test suite, and read both files end-to-end against git show HEAD:... for semantic completeness. Full review at .reviews/latest-review.md.

Test plan

• tests/test-audit-session-load.sh — 9/9 (was 8/8)
• tests/test-doc-consistency.sh — 30/30
• tests/test-docs-usability.sh — 29/29
• tests/test-self-update.sh — 153/153
• tests/test-cli.sh — 78/78
• tests/test-hooks.sh — 154/154
• tests/test-update-skill-step-7-7.sh — 8/8
• tests/test-update-skill-cli-version.sh — 8/8
• tests/test-memory-audit-protocol.sh — 12/12
• All 45 CI unit suites + 4 e2e quick-tests green
• Codex round 1 CERTIFIED 10/10 (no findings)