@dzhechkov/skills-feature-adr 1.3.14 → 1.3.18

package/README.md

@@ -422,9 +422,15 @@ Hooks and events only **collect** raw data — distilling it into reusable patte
   `aqe learning consolidate` periodically (e.g. weekly, or after a heavy QE run);
   `aqe learning stats` shows the experience backlog.
 - **C (dz harness)** — run `dz consolidate` to harvest session learnings into the lexical store
-  (mirrored to the agentdb vector store if present); inspect the result with `dz recall`. The
-  SessionEnd hook automates this **only on clean session ends** — crashed or killed sessions are
-  missed.
+  (mirrored to the agentdb vector store if present); inspect the result with `dz recall`.
+  Consolidation now runs automatically on **PreCompact** — Claude Code fires this before every
+  context compaction (both auto-compaction and manual `/compact`), so long or heavily-compacted
+  sessions consolidate at each compaction boundary rather than only at a clean end. It is
+  **throttled** (a `.dz/.last-consolidate` marker enforces a minimum of ~15 min between auto-runs)
+  so rapid compactions don't reload the embedding model repeatedly. **SessionEnd** remains as a
+  secondary trigger for clean exits, and you can still run `dz consolidate` manually anytime.
+  PreCompact is the primary reliable trigger — earlier the SessionEnd-only hook missed long,
+  compacted, or crashed sessions, which left consolidation effectively manual.
 
 **Degradation symptom:** patterns stop improving while raw logs keep growing. If recall/injection
 quality plateaus, you're collecting but not distilling — consolidate.

package/docs/team-onboarding.md

@@ -216,7 +216,13 @@ reusable patterns is a separate consolidation step:
   returns zero useful patterns. Run `aqe learning consolidate` periodically (weekly, or after a
   heavy QE run); `aqe learning stats` shows the backlog.
 - **C** — run `dz consolidate` (harvests session learnings into the lexical store, mirrored to the
-  vector store if present); inspect with `dz recall`. The SessionEnd hook covers **clean exits only**.
+  vector store if present); inspect with `dz recall`. Consolidation now runs automatically on
+  **PreCompact** (Claude Code fires this before every context compaction — auto and manual
+  `/compact`), so long/compacted sessions consolidate at each compaction boundary, not only at a
+  clean end. It's **throttled** via a `.dz/.last-consolidate` marker (~15 min min between auto-runs).
+  SessionEnd remains a secondary trigger for clean exits; manual `dz consolidate` works anytime.
+  PreCompact is the primary reliable trigger (SessionEnd alone missed long/compacted/crashed
+  sessions).
 
 **Degradation symptom:** patterns stop improving while raw logs keep growing → you're collecting,
 not distilling. Consolidate.

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@dzhechkov/skills-feature-adr",
-  "version": "1.3.14",
+  "version": "1.3.18",
   "description": "Adaptive Feature Development skill pack for Claude Code — 11-step pipeline with Complexity Router (S/M/L/XL), ADR-driven architecture, 15 agentic-qe skills, multi-agent fleet QE. Supports --full-qe, --full-qe-extended, --with-learning, and --knowledge-extractor modes.",
   "bin": {
     "skills-feature-adr": "./bin/cli.js"
@@ -13,9 +13,6 @@
     "CHANGELOG.md",
     "docs/"
   ],
-  "scripts": {
-    "prepack": "node -e \"const fs=require('fs');const bad=['.claude','.skills-feature-adr.json'].filter(p=>fs.existsSync(p));if(bad.length){console.error('prepack guard: stray init artifacts in package dir: '+bad.join(', ')+' — remove before packing');process.exit(1)}\""
-  },
   "keywords": [
     "claude",
     "claude-code",
@@ -60,5 +57,6 @@
   },
   "publishConfig": {
     "access": "public"
-  }
-}
+  },
+  "scripts": {}
+}

package/templates/.claude/skills/feature-adr/SKILL.md

@@ -350,10 +350,14 @@ over agentic-qe's MCP pattern memory (`namespace: "learning"`, `fleet_init` firs
 
 - **Step 0** — `memory_query("patterns/feature-adr/*")` **plus** `dz recall --json` over the durable
   dz store → merged top-3 set `{LEARNED_PATTERNS}`; Step 1 folds them into the requirements brief as advisory lessons.
+  After the recall, record live state:
+  `dz statusline --fa-record --slug <feature-slug> --step "Step 0" --recalled <count of patterns loaded> --stored 0 --mode <reference|direct|direct-extended>`.
 - **Step 8** — recalls `patterns/feature-adr/qe/*` to prime the review checklist; after the gap loop
   closes, stores a `qe-outcome` record in aqe **and** the same lesson via `dz teach` (durable).
+  After the store: `dz statusline --fa-record --slug <slug> --step "Step 8 QE" --recalled <recalled so far> --stored <stored so far>`.
 - **Step 9** — recalls `patterns/feature-adr/fleet/*` to prime agent focus; after the fleet verdict,
   stores a `fleet-qe-finding` in aqe **and** the same lesson via `dz teach`.
+  After the store: `dz statusline --fa-record --slug <slug> --step "Step 9 Fleet" --recalled <recalled so far> --stored <stored so far>` with updated counts.
 
 **Two stores by design:** the aqe kv store is *ephemeral across MCP-server restarts* (janitor/TTL —
 observed live), so it serves fast in-session semantic recall; the dz lexical store
@@ -362,6 +366,20 @@ non-blocking — an error or empty result never stalls the pipeline. This layer
 **distinct from the Keysarium reward layer** installed by `--with-learning`: it uses agentic-qe MCP
 memory + the dz store, is gated on Direct modes only, and never touches `.keysarium/memory/`.
 
+**Live learning panel (`dz statusline`).** The pipeline **drives** the panel: at each pattern-memory-loop
+step above it records its live state via `dz statusline --fa-record …`, so `dz statusline` can surface
+per-run learning (which feature, which step, how many patterns recalled vs. newly stored). Each of the
+Step 0 / Step 8 / Step 9 call-outs above emits one `--fa-record`; the counts accumulate across the run.
+Every checkpoint banner **also** prints a human-visible line so the state shows in the conversation, not
+only the statusline:
+
+```
+🎓 Learning: {recalled} patterns recalled for this feature, {stored} new stored this run
+```
+
+*Honesty note:* the panel is live only insofar as the pipeline records state — it reflects what the
+pipeline actually did with the loop (recalls that ran, stores that landed), not an aspirational count.
+
 ### What changes with `--full-qe`
 
 Full agentic-qe protocols for the same 9 core skills. No new agents, just deeper methodology.
@@ -448,6 +466,7 @@ npx @dzhechkov/skills-feature-adr init --with-learning --knowledge-extractor
 ⏸️ STEP N: [Step Name] Complete
 <promise>[PROMISE_TAG]</promise>
 Tier: {COMPLEXITY_TIER} | Active Steps: {ACTIVE_STEPS}
+🎓 Learning: {recalled} patterns recalled for this feature, {stored} new stored this run
 
 [2-3 line summary]
 Artifacts: [list] ✅