agentic-sdlc-wizard 1.50.0 → 1.51.0

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": "1.50.0",
+      "version": "1.51.0",
       "author": {
         "name": "Stefan Ayala"
       },

package/.claude-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "sdlc-wizard",
-  "version": "1.50.0",
+  "version": "1.51.0",
   "description": "SDLC enforcement for AI agents — TDD, planning, self-review, CI shepherd",
   "author": {
     "name": "Stefan Ayala",

package/CHANGELOG.md

@@ -4,6 +4,28 @@ 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.
 
+## [1.51.0] - 2026-04-27
+
+### Removed
+
+- **`version-test` job in `.github/workflows/weekly-update.yml`** (319 lines) — closes ROADMAP #231 Phase 3a. The CI job ran two-phase E2E (Phase A regression + Phase B improvement) on every detected CC release with `path_to_claude_code_executable` to install + test specific versions. Cost: $8-20/run with zero merged artifacts in 30 days. Replacement: manual local-Max procedure using the v1.49.0+ shepherd (`npm i -g @anthropic-ai/claude-code@<version> && tests/e2e/local-shepherd.sh <PR> --compare-baseline`). Auto-update PRs no longer get auto-validated; the maintainer runs the shepherd before merging.
+
+### Changed
+
+- `tests/test-self-update.sh`: 3 version-test gate tests stubbed to "n/a per #231 Phase 3a" (preserves harness counters; remove after one release if no regressions). Added regression test verifying CI_CD.md documents the manual local replacement.
+- `tests/test-workflow-triggers.sh`: 4 tests updated. Test 86 inverted to ensure version-test stays deleted (any future resurrection emits VERSION_TEST_RESURRECTED). Tests 87, 93, 95 stubbed.
+- `CI_CD.md` "Two-Phase Version Testing" section rewritten with the manual local-shepherd procedure.
+- `plans/AUTO_SELF_UPDATE.md` "Two-Phase Version Testing" section marked DELETED with the local replacement runbook.
+
+### Files
+
+- `.github/workflows/weekly-update.yml` (-319 lines)
+- `tests/test-self-update.sh` (4 tests stubbed/updated)
+- `tests/test-workflow-triggers.sh` (4 tests stubbed/inverted)
+- `CI_CD.md` (1 section rewritten)
+- `plans/AUTO_SELF_UPDATE.md` (1 section rewritten)
+- Version bump 1.50.0 → 1.51.0 across the usual files
+
 ## [1.50.0] - 2026-04-27
 
 ### Added

package/CLAUDE_CODE_SDLC_WIZARD.md

@@ -91,13 +91,20 @@ When your AI tools update, how do you know if the update is safe?
 
 This prevents both false positives (crying wolf) and false negatives (missing real regressions).
 
-**How We Apply This:**
-- Weekly workflow tests new Claude Code versions before recommending upgrade
-- Version-pinned gate: installs the specific CC version and passes it via `path_to_claude_code_executable` so E2E actually runs the new binary
-- Phase A: Does new CC version break SDLC enforcement?
-- Phase B: Do changelog-suggested improvements actually help?
-- Green CI = safe to upgrade. Red = stay on current version until fixed
-- Results shown in PR with statistical confidence
+**How We Apply This (post-ROADMAP #231 Phase 3a, v1.51.0+):**
+- Weekly workflow detects new Claude Code versions and opens an auto-update PR
+- Maintainer runs the version-test locally on Max before merging:
+  ```
+  npm i -g @anthropic-ai/claude-code@<new_version>
+  gh pr checkout <auto_update_pr>
+  tests/e2e/local-shepherd.sh <pr> --compare-baseline
+  ```
+- Phase A semantics (regression): score delta vs main shows whether the new CC version breaks our SDLC enforcement
+- Phase B semantics (improvement): include changelog-suggested doc changes in the PR before running the shepherd
+- Green delta = safe to upgrade. Red = stay on current version until fixed
+- Results posted as a check-run + PR comment with provenance (host, claude version, execution_path)
+
+**Historical:** through v1.50.0, this was a CI cron job (`version-test` in `weekly-update.yml`, $8-20/run) that auto-installed the new version and ran Phase A/B Tier 1+2. Deleted in v1.51.0 because it ran on every release detection regardless of relevance, with zero merged artifacts in 30 days.
 
 ### Benchmark Ceiling Effect (Known Issue — April 2026)
 
@@ -2967,7 +2974,7 @@ If deployment fails or post-deploy verification catches issues:
 
 **SDLC.md:**
 ```markdown
-<!-- SDLC Wizard Version: 1.50.0 -->
+<!-- SDLC Wizard Version: 1.51.0 -->
 <!-- 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] -->
@@ -4032,7 +4039,7 @@ Walk through updates? (y/n)
 Store wizard state in `SDLC.md` as metadata comments (invisible to readers, parseable by Claude):
 
 ```markdown
-<!-- SDLC Wizard Version: 1.50.0 -->
+<!-- SDLC Wizard Version: 1.51.0 -->
 <!-- Setup Date: 2026-01-24 -->
 <!-- Completed Steps: step-0.1, step-0.2, step-1, step-2, step-3, step-4, step-5, step-6, step-7, step-8, step-9 -->
 <!-- Git Workflow: PRs -->

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "agentic-sdlc-wizard",
-  "version": "1.50.0",
+  "version": "1.51.0",
   "description": "SDLC enforcement for Claude Code — hooks, skills, and wizard setup in one command",
   "bin": {
     "sdlc-wizard": "cli/bin/sdlc-wizard.js"

package/skills/update/SKILL.md

@@ -93,9 +93,10 @@ Parse CHANGELOG entries between the user's installed version and latest. Present
 
 ```
 Installed: 1.42.0
-Latest:    1.50.0
+Latest:    1.51.0
 
 What changed:
+- [1.51.0] delete version-test cron (#231 Phase 3a) — manual local-Max replacement via npm i + local-shepherd
 - [1.50.0] local-shepherd.sh --strip-paths flag (#231 Phase 2 — replaces deleted prove-it-test cron)
 - [1.49.0] local-shepherd.sh --compare-baseline flag (#230)
 - [1.48.0] SKILL.md trim — token bloat audit phase 2 follow-up