fix(quick-260507-esn): repair extract_learnings skill ref + drift baseline

Three fixes flushed out by the post-sync CI run:

1. skills/extract_learnings/SKILL.md was still pointing at
   workflows/extract_learnings.md (the old underscore name) after Task 1
   renamed the workflow to extract-learnings.md. Updated both
   <execution_context> and the process delegation line. This was the
   sole Category B (genuinely-missing) drift entry CI flagged.

2. workflows/mvp-phase.md and references/planner-mvp-mode.md (new
   upstream files) used the legacy @~/.claude/get-shit-done/ ref form.
   Rewrote 7 occurrences to the plugin-runtime form
   @${CLAUDE_PLUGIN_ROOT}/ for correct CC ${CLAUDE_PLUGIN_ROOT}
   substitution.

3. Regenerated tests/drift-baseline.json to 131/131/0 (was 123/123/0).
   The +8 increase is the legitimate result of the upstream sync
   adding more plugin-counterpart refs; the drift detector's policy
   is to ratchet baseline when the increase is reviewed (as it is
   here). Category B remains 0.

Author: jnuyens

references/planner-mvp-mode.md

@@ -38,7 +38,7 @@ When the orchestrator sets `WALKING_SKELETON=true` (Phase 1 of a new project und
   - One real DB read/write
   - One real UI interaction wired to the API
   - Deployment to a dev environment (or a documented local-run command that exercises the full stack)
-- The plan **must produce** `SKELETON.md` in the phase directory alongside `PLAN.md`. Use the template at `@~/.claude/get-shit-done/references/skeleton-template.md`. `SKELETON.md` records the architectural decisions that subsequent phases will build on (chosen framework, DB, deployment target, auth approach, directory layout).
+- The plan **must produce** `SKELETON.md` in the phase directory alongside `PLAN.md`. Use the template at `@${CLAUDE_PLUGIN_ROOT}/references/skeleton-template.md`. `SKELETON.md` records the architectural decisions that subsequent phases will build on (chosen framework, DB, deployment target, auth approach, directory layout).
 
 `SKELETON.md` is the architectural backbone for every later vertical slice; treat it as a contract, not a scratchpad.
 

skills/extract_learnings/SKILL.md

@@ -16,7 +16,7 @@ Extract structured learnings from completed phase artifacts (PLAN.md, SUMMARY.md
 </objective>
 
 <execution_context>
-@${CLAUDE_PLUGIN_ROOT}/workflows/extract_learnings.md
+@${CLAUDE_PLUGIN_ROOT}/workflows/extract-learnings.md
 </execution_context>
 
-Execute the extract-learnings workflow from @${CLAUDE_PLUGIN_ROOT}/workflows/extract_learnings.md end-to-end.
+Execute the extract-learnings workflow from @${CLAUDE_PLUGIN_ROOT}/workflows/extract-learnings.md end-to-end.

tests/drift-baseline.json

@@ -1,9 +1,9 @@
 {
   "file_layout": {
-    "total_dangling": 123,
-    "has_plugin_counterpart": 123,
+    "total_dangling": 131,
+    "has_plugin_counterpart": 131,
     "genuinely_missing": 0,
-    "generated_at": "2026-05-01",
+    "generated_at": "2026-05-07",
     "note": "Counts-based ratchet per Phase 7 CONTEXT D-07. Regenerate with --write-baseline when drift is legitimately reduced."
   }
 }

workflows/mvp-phase.md

@@ -3,9 +3,9 @@ Guide the user through MVP-mode planning for a phase. Prompts for an "As a / I w
 </purpose>
 
 <required_reading>
-@~/.claude/get-shit-done/references/user-story-template.md
-@~/.claude/get-shit-done/references/spidr-splitting.md
-@~/.claude/get-shit-done/references/planner-mvp-mode.md
+@${CLAUDE_PLUGIN_ROOT}/references/user-story-template.md
+@${CLAUDE_PLUGIN_ROOT}/references/spidr-splitting.md
+@${CLAUDE_PLUGIN_ROOT}/references/planner-mvp-mode.md
 </required_reading>
 
 <runtime_note>
@@ -29,7 +29,7 @@ Example: /gsd mvp-phase 2.1
 ```
 Exit.
 
-Normalize per `@~/.claude/get-shit-done/references/phase-argument-parsing.md` (zero-pad integer phases to two digits).
+Normalize per `@${CLAUDE_PLUGIN_ROOT}/references/phase-argument-parsing.md` (zero-pad integer phases to two digits).
 
 ## 2. Validate phase exists and check status
 
@@ -73,7 +73,7 @@ Use `AskUserQuestion` with options [Re-prompt / Abort]. On Abort, exit cleanly.
 
 ## 3. User story prompts
 
-Run three sequential `AskUserQuestion` calls. Each is free-text. After all three, assemble into the canonical sentence per `@~/.claude/get-shit-done/references/user-story-template.md`:
+Run three sequential `AskUserQuestion` calls. Each is free-text. After all three, assemble into the canonical sentence per `@${CLAUDE_PLUGIN_ROOT}/references/user-story-template.md`:
 
 **Prompt 1 — As a:**
 > "As a [user role]?"
@@ -112,7 +112,7 @@ If `RE_PROMPT_USER_STORY=true`, re-run only the offending prompt field(s), rebui
 
 ## 4. SPIDR splitting check
 
-Run the SPIDR rules from `@~/.claude/get-shit-done/references/spidr-splitting.md`. Briefly:
+Run the SPIDR rules from `@${CLAUDE_PLUGIN_ROOT}/references/spidr-splitting.md`. Briefly:
 
 **Trigger evaluation.** Check the assembled `USER_STORY` against the four size signals from the reference (compound capabilities, multi-actor, length > 120 chars, vague capability). If none fire, **skip SPIDR** entirely — go to step 5.