agentic-sdlc-wizard 1.82.0 → 1.83.0

package/.claude-plugin/marketplace.json

@@ -13,7 +13,7 @@
       "name": "sdlc-wizard",
       "source": ".",
       "description": "SDLC enforcement for AI agents — TDD, planning, self-review, CI shepherd",
-      "version": "1.82.0",
+      "version": "1.83.0",
       "author": {
         "name": "Stefan Ayala"
       },

package/.claude-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "sdlc-wizard",
-  "version": "1.82.0",
+  "version": "1.83.0",
   "description": "SDLC enforcement for AI agents — TDD, planning, self-review, CI shepherd",
   "author": {
     "name": "Stefan Ayala",

package/CHANGELOG.md

@@ -4,6 +4,17 @@ All notable changes to the SDLC Wizard.
 
 > **Note:** This changelog is for humans to read. Don't manually apply these changes - just run the wizard ("Check for SDLC wizard updates") and it handles everything automatically.
 
+## [1.83.0] - 2026-06-11
+
+### Fixed
+- **#403**: Hook lists multiple wizard-blessed models (claude-opus-4-6, opusplan, fable) instead of hardcoded single-model nudge
+- **#391**: Setup Step 9.5 detects global `[1m]` model pin and warns about billing implications
+- **#405**: Update Step 3 uses `min(npm, CHANGELOG)` as "latest installable" — prevents version race during publish window
+- **#384**: Update Step 7.9 checks effort configuration regardless of version match
+
+### Why
+Model config batch — first trial of the confidence ramp pattern (Opus research → Fable batch review → 95% → TDD → Codex safety check). All 4 issues reached 95% confidence before implementation.
+
 ## [1.82.0] - 2026-06-11
 
 ### Added

package/CLAUDE_CODE_SDLC_WIZARD.md

@@ -3041,7 +3041,7 @@ If deployment fails or post-deploy verification catches issues:
 
 **SDLC.md:**
 ```markdown
-<!-- SDLC Wizard Version: 1.82.0 -->
+<!-- SDLC Wizard Version: 1.83.0 -->
 <!-- Setup Date: [DATE] -->
 <!-- Completed Steps: step-0.1, step-0.2, step-0.4, step-1, step-2, step-3, step-4, step-5, step-6, step-7, step-8, step-9 -->
 <!-- Git Workflow: [PRs or Solo] -->

package/hooks/model-effort-check.sh

@@ -11,7 +11,8 @@
 #
 # Non-blocking: always exits 0.
 
-RECOMMENDED_MODEL="claude-opus-4-6"
+# Multiple wizard-blessed models — don't nudge to a single one (#403)
+RECOMMENDED_MODELS="claude-opus-4-6, opusplan, or fable (run: /model)"
 
 HOOK_DIR="${BASH_SOURCE[0]%/*}"
 [ "$HOOK_DIR" = "${BASH_SOURCE[0]}" ] && HOOK_DIR="."
@@ -63,7 +64,7 @@ echo ""
 echo " Run: /effort max"
 echo " Persist: set CLAUDE_CODE_EFFORT_LEVEL=max in settings env block"
 echo ""
-echo " recommended model: $RECOMMENDED_MODEL (run: /model $RECOMMENDED_MODEL)"
+echo " recommended models: $RECOMMENDED_MODELS"
 echo "=============================================================================="
 
 exit 0

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "agentic-sdlc-wizard",
-  "version": "1.82.0",
+  "version": "1.83.0",
   "description": "SDLC enforcement for Claude Code — hooks, skills, and wizard setup in one command",
   "bin": {
     "sdlc-wizard": "cli/bin/sdlc-wizard.js"

package/skills/setup/SKILL.md

@@ -232,6 +232,8 @@ The CLI ships `cli/templates/settings.json` with **no** `model` or `env` pin by
 
 **Why this is opt-in (issue #198):** A top-level `"model"` in `settings.json` disables auto-mode for the session. Pinning is only worth it when you want consistent model behavior (opusplan or flagship) rather than per-turn auto-selection.
 
+**Check for global `[1m]` model pin (#391):** Read `~/.claude/settings.json`. If `model` contains `[1m]`, warn: global pin forces 1M on every repo; headless surfaces bill against credits after June 15. Offer `[r] Remove / [k] Keep`. If `r`: delete the `model` key.
+
 **Run the complexity heuristic first (roadmap #233):**
 
 ```bash

package/skills/update/SKILL.md

@@ -89,13 +89,16 @@ Extract latest version from the first `## [X.X.X]` line.
 
 ### Step 3: Compare Versions and Show What Changed
 
-Parse CHANGELOG entries between the user's installed version and latest. Present a clear summary:
+**Resolve "latest installable" from npm registry (#405):** Compare the npm registry version (Step 1.5 cache) to the CHANGELOG heading version (Step 2). Use the **lower** of the two as "latest installable" — this prevents showing a version that isn't yet published to npm during the publish window. If CHANGELOG is ahead of npm, note: "v{changelog} is on GitHub but not yet published to npm; showing v{npm} as latest installable."
+
+Parse CHANGELOG entries between the user's installed version and the resolved latest installable. Present a clear summary:
 
 ```
 Installed: 1.42.0
-Latest:    1.82.0
+Latest:    1.83.0
 
 What changed:
+- [1.83.0] Model config batch: multi-model hook recommendation (#403), global [1m] pin detection (#391), version race fix (#405), effort config check (#384).
 - [1.82.0] Usage diagnostics: fix /usage row, Reading Usage Signals guide, advisor fallback procedure, Fable effort guidance, autocompact cross-reference.
 - [1.81.0] Native `advisorModel` support: Setup A gets Fable advisor, Setup B gets Opus advisor. Replaces manual subagent spawning. Requires CC v2.1.170+.
 - [1.80.0] Flip default: Opus 4.6 max becomes recommended flagship; Opus 4.8 demoted to opt-in `[l] Latest` tier.
@@ -126,8 +129,8 @@ Read the actual entries from the fetched CHANGELOG; don't paraphrase. The user w
 
 **If versions match:** Step 7.7 (global plugin-registration cleanup) is independent of wizard file versions — it must run even when the user is up-to-date. The `check-only` flag still gates whether cleanup is *applied*:
 
-- **Without `check-only`**: Run Step 7.7 in normal mode (detect, prompt, apply) before stopping. Then say "You're up to date! (version X.X.X)" and stop. Do not run Steps 4–10; only Step 7.7 fires on match.
-- **With `check-only`**: Run Step 7.7 in detection-only mode — report any dead plugin registrations, but do NOT prompt and do NOT mutate `~/.claude/settings.json`. Then say "You're up to date! (version X.X.X)" and stop.
+- **Without `check-only`**: Run Step 7.7 and Step 7.9 in normal mode (detect, prompt, apply) before stopping. Then say "You're up to date! (version X.X.X)" and stop. Do not run Steps 4–10; only Steps 7.7 and 7.9 fire on match.
+- **With `check-only`**: Run Step 7.7 and Step 7.9 in detection-only mode — report findings but do NOT prompt and do NOT mutate settings. Then say "You're up to date! (version X.X.X)" and stop.
 
 **If user passed `check-only` and versions don't match:** Stop after showing what changed. Do not apply anything.
 
@@ -236,6 +239,15 @@ If CC < v2.1.170: show "Run `! claude update` to upgrade" and skip. If `.claude/
 
 If `model` pin exists but no `advisorModel`, suggest: `claude-opus-4-6`/`claude-opus-4-8` → `advisorModel: "fable"`, `opusplan` → `advisorModel: "claude-opus-4-6"`. Ask `[a/S]`, write only `advisorModel` if accepted.
 
+### Step 7.9: Effort Configuration Check (#384)
+
+Runs regardless of version match (like Step 7.7). In `check-only` mode: report only, no mutations.
+
+1. Check `CLAUDE_CODE_EFFORT_LEVEL` env var. If set to `max`: **pass** (silent — this is the correct persistence path).
+2. If env var is unset, check `effortLevel` in settings cascade (`.claude/settings.local.json` → `.claude/settings.json` → `~/.claude/settings.json`).
+3. If `effortLevel` is `max` but `CLAUDE_CODE_EFFORT_LEVEL` env var is unset: warn that CC ignores session-only max in settings — only the env var persists it. Suggest adding to the `env` block in `.claude/settings.json`.
+4. If effort is unset or below `max`: warn "SDLC requires max effort" and suggest `/effort max` + env var persistence.
+
 ### Step 8: Apply Selected Changes
 
 For each approved file: Edit (existing) or Write (MISSING). For settings.json, apply the merge from Step 7.