agentic-qe 3.7.11 → 3.7.13

package/.claude/skills/.validation/schemas/skill-frontmatter.schema.json

@@ -226,6 +226,11 @@
       "default": 0,
       "description": "Trust tier level for validation. 0=Declarative (SKILL.md only), 1=Structured (+ JSON output schema), 2=Validated (+ executable validator), 3=Evaluated (+ evaluation test suite)"
     },
+    "skill_intent": {
+      "type": "string",
+      "enum": ["capability_uplift", "encoded_preference", "hybrid"],
+      "description": "Skill intent classification. 'capability_uplift' = fills gaps in base model abilities, needs regression testing as models improve. 'encoded_preference' = automates established workflows/standards, needs fidelity verification. 'hybrid' = combines both intents."
+    },
     "validation": {
       "type": "object",
       "description": "Validation configuration for Trust But Verify system (required for trust_tier >= 1)",

package/.claude/skills/release/SKILL.md

@@ -195,7 +195,23 @@ node /workspaces/agentic-qe-new/dist/cli/bundle.js health 2>&1 | head -10
 ```
 These should respond (even if empty results) without errors, confirming the subsystems initialize properly.
 
-#### 8e. Cleanup
+#### 8e. Isolated Dependency Check (catches missing externals)
+```bash
+# Pack and install in a clean temp directory to simulate real user install
+CLEAN_DIR=$(mktemp -d)
+npm pack --pack-destination "$CLEAN_DIR" 2>&1 | tail -2
+cd "$CLEAN_DIR"
+npm init -y > /dev/null 2>&1
+npm install ./agentic-qe-<version>.tgz 2>&1 | tail -3
+node node_modules/.bin/aqe --version 2>&1
+EXIT=$?
+echo "Exit code: $EXIT"
+cd /workspaces/agentic-qe-new
+rm -rf "$CLEAN_DIR"
+```
+Must exit 0 and print the correct version. If it crashes with `ERR_MODULE_NOT_FOUND`, a dependency is marked as external in the build scripts but not listed in `dependencies`. Fix by either bundling it, lazy-loading it, or adding it to dependencies.
+
+#### 8f. Cleanup
 ```bash
 rm -rf /tmp/aqe-release-test
 ```
@@ -328,11 +344,37 @@ gh run view <run-id> --log-failed
 ```bash
 npm view agentic-qe@<version> name version
 ```
-Confirm the published version matches. Test install:
+Confirm the published version matches. Test install in local environment:
 ```bash
 npx agentic-qe@<version> --version
 ```
 
+### 15. Isolated Install Verification (CRITICAL)
+
+This step catches missing/external dependency issues that only manifest in clean environments (e.g., `typescript` not being available when installed globally). This MUST pass before declaring the release successful.
+
+```bash
+# Create a completely isolated install — no access to project node_modules
+CLEAN_DIR=$(mktemp -d)
+npm install --prefix "$CLEAN_DIR" agentic-qe@<version> 2>&1 | tail -5
+
+# Test CLI commands using ONLY the isolated install's dependencies
+NODE_PATH="$CLEAN_DIR/node_modules" node "$CLEAN_DIR/node_modules/.bin/aqe" --version
+NODE_PATH="$CLEAN_DIR/node_modules" node "$CLEAN_DIR/node_modules/.bin/aqe" --help 2>&1 | head -5
+
+# Cleanup
+rm -rf "$CLEAN_DIR"
+```
+
+If `--version` crashes (e.g., `ERR_MODULE_NOT_FOUND`), the release has a broken dependency. Diagnose whether the missing package should be:
+- **Bundled** into the CLI (add to build script, remove from externals)
+- **Added to `dependencies`** in package.json (if it's a real runtime dep)
+- **Lazy-loaded** with try/catch (if only needed for optional features)
+
+Fix, rebuild, and re-release if this step fails. Never ship a CLI that crashes on `--version`.
+
+**STOP — confirm isolated install works.**
+
 ## Rules
 
 - **Single package.json** at root — no v3/ subdirectory exists

package/.claude/skills/skills-manifest.json

@@ -904,7 +904,7 @@
   },
   "metadata": {
     "generatedBy": "Agentic QE Fleet",
-    "fleetVersion": "3.7.11",
+    "fleetVersion": "3.7.13",
     "manifestVersion": "1.3.0",
     "lastUpdated": "2026-02-04T00:00:00.000Z",
     "contributors": [

package/CHANGELOG.md

@@ -5,6 +5,26 @@ All notable changes to the Agentic QE project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [3.7.13] - 2026-03-07
+
+### Added
+
+- **Trigger Optimizer** — Analyzes skill descriptions and tags to detect false positive/negative activation risks. Calculates Jaccard similarity across the skill fleet, identifies confusable skills, and generates actionable suggestions to improve trigger precision.
+- **Version Comparator** — A/B testing between skill versions using Cohen's d effect size and confidence scoring. Produces per-test-case comparisons and Markdown reports for data-driven skill improvement decisions.
+- **Skill Intent Classification** — New `skill_intent` frontmatter field classifies skills as `capability_uplift` (fills model gaps), `encoded_preference` (encodes team workflows), or `hybrid`. Drives different validation strategies per intent type.
+
+### Changed
+
+- ADR-056 updated with Phase 6 (Blog-Inspired Improvements) documenting the three new features.
+- Validation module index now exports TriggerOptimizer, VersionComparator, and all associated types.
+
+## [3.7.12] - 2026-03-06
+
+### Fixed
+
+- **CLI crash on global install** — `aqe --version` crashed with `ERR_MODULE_NOT_FOUND: Cannot find package 'typescript'` when installed globally (`npm i -g agentic-qe`). TypeScript was marked as an external ESM dependency in the build but is only a devDependency, so it's absent in clean environments. Now lazy-loaded via `createRequire` Proxy — only loads when AST parsing features are actually used.
+- **Release skill missing isolated install check** — Added pre-release step 8e (isolated dependency check) and post-publish step 15 (clean-environment install verification) to catch missing external dependencies before they reach users.
+
 ## [3.7.11] - 2026-03-06
 
 ### Added

package/assets/skills/.validation/schemas/skill-frontmatter.schema.json

@@ -226,6 +226,11 @@
       "default": 0,
       "description": "Trust tier level for validation. 0=Declarative (SKILL.md only), 1=Structured (+ JSON output schema), 2=Validated (+ executable validator), 3=Evaluated (+ evaluation test suite)"
     },
+    "skill_intent": {
+      "type": "string",
+      "enum": ["capability_uplift", "encoded_preference", "hybrid"],
+      "description": "Skill intent classification. 'capability_uplift' = fills gaps in base model abilities, needs regression testing as models improve. 'encoded_preference' = automates established workflows/standards, needs fidelity verification. 'hybrid' = combines both intents."
+    },
     "validation": {
       "type": "object",
       "description": "Validation configuration for Trust But Verify system (required for trust_tier >= 1)",