xtrm-tools 0.7.17 → 0.7.19

package/cli/package.json

@@ -1,6 +1,7 @@
 {
   "name": "xtrm-cli",
-  "version": "0.7.17",
+  "private": true,
+  "version": "0.7.19",
   "description": "Claude Code tools installer (skills, hooks, MCP servers)",
   "main": "./dist/index.js",
   "type": "module",
@@ -36,14 +37,14 @@
     "ora": "^9.3.0",
     "project": "^0.1.6",
     "prompts": "^2.4.2",
-    "tdd-guard": "^1.1.0",
+    "yaml": "^2.4.2",
     "zod": "^4.3.6"
   },
   "devDependencies": {
     "@types/fs-extra": "^11.0.4",
     "@types/node": "^25.3.0",
+    "@types/prompts": "^2.4.9",
     "fast-check": "^4.6.0",
-    "tdd-guard-vitest": "^0.1.6",
     "tsup": "^8.5.1",
     "tsx": "^4.21.0",
     "typescript": "^5.9.3",
@@ -51,5 +52,10 @@
   },
   "engines": {
     "node": ">=20.0.0"
+  },
+  "pnpm": {
+    "overrides": {
+      "vite": "8.0.12"
+    }
   }
 }

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "xtrm-tools",
-  "version": "0.7.17",
+  "version": "0.7.19",
   "description": "Claude Code tools installer (skills, hooks, MCP servers)",
   "license": "MIT",
   "type": "module",
@@ -32,7 +32,10 @@
     "!packages/pi-extensions/extensions/**/.pi/**",
     "!.xtrm/extensions/**/.pi/**",
     "!.xtrm/**/test_*.py",
-    "!.xtrm/**/tests/**"
+    "!.xtrm/**/tests/**",
+    "!.xtrm/skills/default/**/evals/**",
+    "!.xtrm/skills/default/**/workspace/iteration-*/**",
+    "!packages/*/.serena/**"
   ],
   "publishConfig": {
     "access": "public"
@@ -56,25 +59,37 @@
     "start": "node cli/dist/index.cjs",
     "lint": "echo 'No linting configured'",
     "test": "npm test --workspace cli",
+    "check:package": "node scripts/assert-no-self-dependency.mjs",
     "version": "npm run sync:cli-version && git add cli/package.json packages/pi-extensions/package.json",
-    "prepublishOnly": "npm run sync:cli-version && node scripts/vendor-specialists-skills.mjs && npm run gen-registry && npm run build",
+    "check:skills-ownership": "node scripts/check-skills-ownership.mjs",
+    "check:specialists-vendor": "node scripts/verify-specialists-vendor.mjs",
+    "check:asset-contract": "node scripts/verify-asset-contract.mjs",
+    "check:layout-guards": "node scripts/check-layout-guards.mjs",
+    "check:gitnexus-no-counter": "node scripts/check-gitnexus-no-counter.mjs",
+    "check:skills-symlinks": "node scripts/check-skills-symlinks.mjs",
+    "prepublishOnly": "npm run sync:cli-version && npm run check:package && npm run check:skills-ownership && node scripts/vendor-specialists-skills.mjs --specialists-package specialists --specialists-ref master && npm run gen-registry && npm run check:registry-pack-parity && npm run check:payload-hygiene && npm run check:specialists-vendor && npm run check:layout-guards && npm run check:gitnexus-no-counter && npm run check:skills-symlinks && npm run build",
     "release": "npm publish --tag latest",
     "release:pi-extensions": "npm publish --workspace @jaggerxtrm/pi-extensions --tag latest --access public",
     "release:all": "npm run release && npm run release:pi-extensions",
-    "gen-registry": "node scripts/gen-registry.mjs"
+    "gen-registry": "node scripts/gen-registry.mjs",
+    "check:registry-pack-parity": "node scripts/check-registry-pack-parity.mjs",
+    "check:payload-hygiene": "node scripts/check-payload-hygiene.mjs"
   },
   "engines": {
     "node": ">=20.0.0"
   },
   "dependencies": {
-    "@artale/pi-procs": "^1.1.0",
     "comment-json": "^4.2.3",
     "conf": "^15.1.0",
     "dotenv": "^17.3.1",
     "fs-extra": "^11.2.0",
     "kleur": "^4.1.5",
-    "prompts": "^2.4.2",
-    "xtrm-tools": "^2.0.2"
+    "prompts": "^2.4.2"
+  },
+  "overrides": {
+    "fast-uri": "3.1.2",
+    "@aws-sdk/xml-builder": "3.972.22",
+    "vite": "8.0.12"
   },
   "pi": {
     "extensions": [
@@ -88,5 +103,10 @@
       "npm:@aliou/pi-guardrails",
       "npm:@aliou/pi-processes"
     ]
+  },
+  "pnpm": {
+    "overrides": {
+      "vite": "8.0.12"
+    }
   }
 }

package/packages/pi-extensions/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@jaggerxtrm/pi-extensions",
-  "version": "0.7.17",
+  "version": "0.7.19",
   "description": "Unified Pi extension entrypoint for xtrm-managed extensions",
   "type": "module",
   "publishConfig": {

package/.xtrm/skills/default/planning/evals/evals.json

@@ -1,19 +0,0 @@
-{
-  "skill_name": "planning",
-  "evals": [
-    {
-      "id": 1,
-      "eval_name": "docs-list-command",
-      "prompt": "Plan the implementation of the `xtrm docs list` command (xtrm-vwp0). The command should list all project docs with metadata, support filtering, table output, and JSON mode. It needs to be a subcommand of the existing `xtrm docs` CLI group in cli/src/. There's already a partially-implemented docs.ts somewhere. Break this into a proper phased issue board.",
-      "expected_output": "An epic with phased child tasks (scaffold/core/integration), each with rich descriptions containing context, what to build, AC, and approach notes. test-planning invoked after issue board created. Dependencies wired between phases.",
-      "files": []
-    },
-    {
-      "id": 2,
-      "eval_name": "docs-crosscheck-command",
-      "prompt": "Plan the implementation of the `xtrm docs cross-check` command (xtrm-uc0e). This validates docs against PRs and bd issues — detects stale docs, coverage gaps, open issue references. Uses gh CLI for GitHub data. Needs to be a subcommand of `xtrm docs`. Break this into a well-structured bd issue board with proper phasing.",
-      "expected_output": "An epic with phased child tasks covering: GitHub data fetching (boundary layer), cross-check logic (core layer), CLI command wiring (shell layer). test-planning invoked. High-quality issue descriptions that another agent could work from independently.",
-      "files": []
-    }
-  ]
-}

package/.xtrm/skills/default/quality-gates/evals/evals.json

@@ -1,181 +0,0 @@
-{
-  "skill_name": "using-quality-gates",
-  "evals": [
-    {
-      "id": 1,
-      "name": "typescript-feature-with-tests",
-      "prompt": "Add a new authentication middleware to src/auth.ts that validates JWT tokens. The middleware should check the Authorization header, verify the token, and attach the user to the request object.",
-      "expected_output": "Skill explains TDD workflow: write failing test first, then implement. References tdd-guard-vitest or tdd-guard-jest setup. After implementation, mentions TS quality gate will run ESLint/Prettier checks.",
-      "expectations": [
-        "Mentions writing a failing test before implementation",
-        "References TDD Guard blocking mechanism",
-        "Mentions TypeScript quality gate runs after edit",
-        "Provides actionable next steps"
-      ]
-    },
-    {
-      "id": 2,
-      "name": "python-refactor-request",
-      "prompt": "Refactor the database connection pooling in db/connection.py to use async/await. Current implementation is blocking and causing performance issues.",
-      "expected_output": "Skill explains TDD + Python quality workflow: write failing test for async behavior, implement, then ruff/mypy will validate. Mentions auto-fix capabilities.",
-      "expectations": [
-        "Mentions writing tests first (TDD Guard)",
-        "References Python quality gate (ruff + mypy)",
-        "Mentions auto-fix for linting issues",
-        "Explains the post-edit validation flow"
-      ]
-    },
-    {
-      "id": 3,
-      "name": "quality-gate-error-fix",
-      "prompt": "I'm getting blocked by the quality gate with TypeScript errors. Here's the error: 'Type string is not assignable to type number'. How do I fix this?",
-      "expected_output": "Skill explains quality gate error handling: read errors, apply auto-fix if available, manually fix type errors, gate re-runs automatically.",
-      "expectations": [
-        "Explains how to read quality gate errors",
-        "Mentions auto-fix capability",
-        "Explains manual fix process for type errors",
-        "Notes gate re-runs on next edit"
-      ]
-    },
-    {
-      "id": 4,
-      "name": "partial-install-python-only",
-      "prompt": "I'm working on a Python-only project with pytest. What quality tools should I install?",
-      "expected_output": "Skill recommends Python-specific setup: tdd-guard-pytest for TDD, ruff + mypy for quality gate. Explains partial install workflow.",
-      "expectations": [
-        "Recommends tdd-guard-pytest",
-        "Recommends ruff and mypy",
-        "Explains Python-only workflow",
-        "Does not mention TypeScript tools"
-      ]
-    },
-    {
-      "id": 5,
-      "name": "tdd-guard-blocking-confusion",
-      "prompt": "Why am I getting 'No failing test found' when I try to edit src/service.ts? I just want to add a logging statement.",
-      "expected_output": "Skill explains TDD Guard purpose: enforce test-first development. Even small changes require a failing test. Suggests writing a test that verifies the logging behavior.",
-      "expectations": [
-        "Explains TDD Guard blocks all implementation",
-        "Clarifies test-first requirement",
-        "Suggests writing appropriate test",
-        "Does not suggest bypassing the guard"
-      ]
-    },
-    {
-      "id": 6,
-      "name": "eslint-not-found-error",
-      "prompt": "The quality gate says 'ESLint not found' but I'm editing a TypeScript file. What do I do?",
-      "expected_output": "Skill explains ESLint is required for TS quality gate. Provides install command (npm install --save-dev eslint) or how to disable in hook-config.json.",
-      "expectations": [
-        "Explains ESLint is required dependency",
-        "Provides npm install command",
-        "Mentions hook-config.json disable option",
-        "Clear troubleshooting steps"
-      ]
-    },
-    {
-      "id": 7,
-      "name": "full-workflow-question",
-      "prompt": "Walk me through the complete workflow for adding a new feature to this TypeScript project.",
-      "expected_output": "Skill explains full pipeline: 1) Write failing test, 2) TDD Guard allows implementation, 3) Implement feature, 4) TS quality gate validates, 5) Fix any issues, 6) Commit.",
-      "expectations": [
-        "Lists all steps in order",
-        "Explains TDD Guard role",
-        "Explains quality gate role",
-        "Mentions auto-fix capabilities",
-        "Includes commit step"
-      ]
-    },
-    {
-      "id": 8,
-      "name": "documentation-edit-exception",
-      "prompt": "I need to update the README.md with new API documentation. Will the quality gates block me?",
-      "expected_output": "Skill explains quality gates only apply to code files (.ts, .js, .py, etc.), not documentation. README edits proceed without TDD or linting checks.",
-      "expectations": [
-        "Clarifies documentation is exempt",
-        "Lists code file extensions that trigger gates",
-        "Explains why docs are exempt",
-        "No TDD requirement for docs"
-      ]
-    },
-    {
-      "id": 9,
-      "name": "mypy-type-errors",
-      "prompt": "Mypy is reporting 15 type errors in my Python code. Should I fix all of them before continuing?",
-      "expected_output": "Skill explains exit code 2 means blocking - all errors must be fixed. Suggests using mypy's output to prioritize, fix incrementally, gate re-runs on each edit.",
-      "expectations": [
-        "Confirms all blocking errors must be fixed",
-        "Explains exit code 2 behavior",
-        "Suggests incremental fix approach",
-        "Notes gate re-runs automatically"
-      ]
-    },
-    {
-      "id": 10,
-      "name": "vitest-reporter-setup",
-      "prompt": "How do I set up the Vitest test reporter for TDD Guard?",
-      "expected_output": "Skill explains tdd-guard-vitest installation and vitest.config.ts configuration with VitestReporter and project root path.",
-      "expectations": [
-        "Mentions tdd-guard-vitest package",
-        "Shows vitest.config.ts configuration",
-        "Explains project root path requirement",
-        "Clear setup steps"
-      ]
-    },
-    {
-      "id": 11,
-      "name": "should-not-trigger-general-chat",
-      "prompt": "What's the weather like today?",
-      "expected_output": "Skill should NOT trigger - this is general chat, not a code quality workflow question.",
-      "expectations": [
-        "Skill does not activate",
-        "Standard Claude response"
-      ]
-    },
-    {
-      "id": 12,
-      "name": "should-not-trigger-unrelated-coding",
-      "prompt": "Write a Python script to scrape data from example.com and save it to CSV.",
-      "expected_output": "Skill should NOT trigger strongly - this is a general coding task without quality gate context. May mention TDD as best practice but full skill not needed.",
-      "expectations": [
-        "Minimal or no skill activation",
-        "Focus on task completion",
-        "May mention testing as best practice"
-      ]
-    },
-    {
-      "id": 13,
-      "name": "should-not-trigger-math-question",
-      "prompt": "What's the time complexity of binary search?",
-      "expected_output": "Skill should NOT trigger - this is a CS theory question, not about quality gates.",
-      "expectations": [
-        "Skill does not activate",
-        "Standard Claude response"
-      ]
-    },
-    {
-      "id": 14,
-      "name": "edge-case-mixed-language-project",
-      "prompt": "I have a monorepo with both TypeScript backend and Python ML services. How do quality gates work?",
-      "expected_output": "Skill explains both gates can coexist: TS quality gate for backend files, PY quality gate for ML services. TDD Guard works with both via appropriate reporters.",
-      "expectations": [
-        "Explains coexistence of both gates",
-        "File-type-based routing",
-        "TDD Guard works with both",
-        "Separate reporters per language"
-      ]
-    },
-    {
-      "id": 15,
-      "name": "edge-case-auto-fix-verification",
-      "prompt": "The quality gate said it auto-fixed 3 issues but I still have 2 errors. What happened?",
-      "expected_output": "Skill explains auto-fix handles fixable issues (formatting, simple lint), but type errors and complex issues require manual fixes. Shows how to identify remaining issues.",
-      "expectations": [
-        "Explains auto-fix limitations",
-        "Distinguishes fixable vs manual issues",
-        "Type errors require manual fix",
-        "How to read remaining errors"
-      ]
-    }
-  ]
-}

package/.xtrm/skills/default/quality-gates/workspace/iteration-1/FINAL-EVAL-SUMMARY.md

@@ -1,75 +0,0 @@
-# Using Quality Gates — Skill Creator Evals (Iteration 1) — COMPLETE
-
-## All 10 Test Cases Evaluated
-
-### Should-Trigger Eval Results
-
-| ID | Name | Expectations Met | Notes |
-|----|------|------------------|-------|
-| 1 | typescript-feature-with-tests | ✅ 4/4 | Full TDD + TS workflow |
-| 2 | python-refactor-request | ✅ 4/4 | Async refactor + PY gate |
-| 3 | quality-gate-error-fix | ✅ 4/4 | Error handling explained |
-| 4 | partial-install-python-only | ✅ 4/4 | Python-only, no TS mentions |
-| 5 | tdd-guard-blocking-confusion | ✅ 4/4 | TDD philosophy explained |
-| 14 | edge-case-mixed-language-project | ✅ 4/4 | Coexistence explained |
-| 15 | edge-case-auto-fix-verification | ✅ 4/4 | Auto-fix limits clarified |
-
-### Should-NOT-Trigger Eval Results
-
-| ID | Name | Result | Notes |
-|----|------|--------|-------|
-| 11 | should-not-trigger-general-chat | ✅ Pass | Skill correctly silent |
-| 12 | should-not-trigger-unrelated-coding | ✅ Pass | Minimal mode applied |
-| 13 | should-not-trigger-math-question | ✅ Pass | Skill correctly silent |
-
----
-
-## Overall Assessment
-
-### Pass Rate: 10/10 (100%) ✅
-
-**Iteration 1 Fix Applied:**
-- Added "Response Modes" section to SKILL.md
-- Full Workflow Mode for feature/refactor work
-- Minimal Mode for general coding tasks
-- Eval 12 re-run verified fix works
-
----
-
-## Skill Files Created
-
-```
-project-skills/quality-gates/
-├── README.md                          # User documentation
-├── evals/
-│   └── evals.json                     # 15 test cases defined
-├── .claude/
-│   └── skills/
-│       └── using-quality-gates/
-│           └── SKILL.md               # Main skill file (with Response Modes)
-└── workspace/
-    └── iteration-1/
-        ├── [10 eval directories with outputs]
-        └── FINAL-EVAL-SUMMARY.md
-```
-
----
-
-## Ready to Ship
-
-The skill is complete and passes all evals. Key features:
-
-1. **Unified workflow** — TDD Guard + TS/PY Quality Gates in one skill
-2. **Progressive disclosure** — Full vs Minimal mode based on context
-3. **Language routing** — TypeScript vs Python handled correctly
-4. **Partial install support** — Works with any combination of gates
-5. **Troubleshooting** — Clear guidance for common errors
-
----
-
-## Next Steps
-
-1. ✅ Skill created and validated
-2. ⏳ Test installation: `xtrm install project quality-gates`
-3. ⏳ Update xtrm CLI to include quality-gates in project list
-4. ⏳ Decide: Deprecate individual skills or keep as legacy options

package/.xtrm/skills/default/quality-gates/workspace/iteration-1/edge-case-auto-fix-verification/with_skill/outputs/response.md

@@ -1,59 +0,0 @@
-# Eval 15: edge-case-auto-fix-verification
-
-**Prompt:** The quality gate said it auto-fixed 3 issues but I still have 2 errors. What happened?
-
-**Skill Applied:** using-quality-gates
-
----
-
-## Response (with skill active)
-
-This is expected behavior. Here's what happened:
-
-### Auto-Fix Limitations
-
-**Auto-fix handles:**
-- ✅ Prettier formatting issues (indentation, quotes, semicolons)
-- ✅ ESLint stylistic rules (unused vars, prefer const, etc.)
-- ✅ Ruff linting issues (imports, style, simple fixes)
-
-**Auto-fix CANNOT handle:**
-- ❌ TypeScript type errors (wrong types, missing properties)
-- ❌ Mypy type errors (type mismatches)
-- ❌ Complex logic errors
-- ❌ Missing imports or undefined variables
-
-### Reading the Output
-
-```
-[WARN] Auto-fix applied: 3 issues fixed
-[ERROR] TypeScript compilation failed:
-  src/user.ts:42:5 - error TS2322: Type 'string' is not assignable to type 'number'
-[ERROR] ESLint found 1 issue that could not be auto-fixed:
-  src/user.ts:28:3 - Missing return type on function
-[BLOCK] 2 issues remain - fix before continuing
-```
-
-### How to Fix Remaining Issues
-
-1. **Type errors** — Add proper types or convert values:
-   ```typescript
-   // Fix: convert string to number
-   const userId: number = Number(inputId);
-   ```
-
-2. **Missing return types** — Add explicit type annotations:
-   ```typescript
-   // Fix: add return type
-   function getUser(id: number): User { ... }
-   ```
-
-3. **Save the file** — Quality Gate re-runs automatically on next edit
-
-### Why This Design?
-
-Auto-fix is conservative — it only applies fixes that are guaranteed safe. Type errors require human judgment to fix correctly.
-
----
-
-**Result:** ✅ Explains auto-fix limitations, distinguishes fixable vs manual issues

package/.xtrm/skills/default/quality-gates/workspace/iteration-1/edge-case-mixed-language-project/with_skill/outputs/response.md

@@ -1,60 +0,0 @@
-# Eval 14: edge-case-mixed-language-project
-
-**Prompt:** I have a monorepo with both TypeScript backend and Python ML services. How do quality gates work?
-
-**Skill Applied:** using-quality-gates
-
----
-
-## Response (with skill active)
-
-In a **mixed-language monorepo**, Quality Gates work independently per file type:
-
-### File-Type Routing
-
-| File Extension | Active Gate |
-|----------------|-------------|
-| `.ts`, `.tsx`, `.js`, `.jsx` | TypeScript Quality Gate |
-| `.py` | Python Quality Gate |
-| `.go`, `.rs`, `.php` | TDD Guard only (no language-specific quality gate) |
-| `.md`, `.txt`, `.json` | No gate (documentation/config) |
-
-### TDD Guard (All Code Files)
-
-TDD Guard applies to **all code files** regardless of language:
-- TypeScript backend tests → `tdd-guard-jest` or `tdd-guard-vitest`
-- Python ML tests → `tdd-guard-pytest`
-- Both reporters can coexist in the same project
-
-### Quality Gates (Language-Specific)
-
-**TypeScript Backend** (`packages/backend/`):
-```bash
-npm install --save-dev typescript eslint prettier
-# PostToolUse hook runs: tsc + eslint + prettier
-```
-
-**Python ML Services** (`packages/ml-service/`):
-```bash
-pip install ruff mypy
-# PostToolUse hook runs: ruff check + ruff format + mypy
-```
-
-### Configuration
-
-Each language has separate config:
-- TS: `.claude/hooks/hook-config.json`
-- Python: Environment variables (`CLAUDE_HOOKS_RUFF_ENABLED`, etc.)
-
-### Your Workflow
-
-```
-Edit TypeScript file → TDD Guard check → TS Quality Gate
-Edit Python file     → TDD Guard check → PY Quality Gate
-```
-
-Both gates coexist peacefully — each only activates for its file types.
-
----
-
-**Result:** ✅ Explains coexistence, file-type routing, separate reporters

package/.xtrm/skills/default/quality-gates/workspace/iteration-1/eval-summary.md

@@ -1,105 +0,0 @@
-# Using Quality Gates — Skill Creator Evals (Iteration 1)
-
-## Test Cases Run: 5 of 15
-
-### Eval Results Summary
-
-| ID | Name | Status | Notes |
-|----|------|--------|-------|
-| 1 | typescript-feature-with-tests | ✅ Complete | Full TDD + TS quality gate workflow explained |
-| 2 | python-refactor-request | ✅ Complete | Async refactor with Python quality gate |
-| 3 | quality-gate-error-fix | ✅ Complete | Error handling and auto-fix explanation |
-| 4 | partial-install-python-only | ✅ Complete | Python-only setup, no TS tools mentioned |
-| 5 | tdd-guard-blocking-confusion | ✅ Complete | Explains TDD philosophy, provides options |
-
----
-
-## Evaluation Criteria (from evals/evals.json)
-
-### Eval 1: typescript-feature-with-tests
-**Expectations:**
-- [ ] Mentions writing a failing test before implementation
-- [ ] References TDD Guard blocking mechanism
-- [ ] Mentions TypeScript quality gate runs after edit
-- [ ] Provides actionable next steps
-
-**Result:** ✅ All expectations met
-
----
-
-### Eval 2: python-refactor-request
-**Expectations:**
-- [ ] Mentions writing tests first (TDD Guard)
-- [ ] References Python quality gate (ruff + mypy)
-- [ ] Mentions auto-fix for linting issues
-- [ ] Explains the post-edit validation flow
-
-**Result:** ✅ All expectations met
-
----
-
-### Eval 3: quality-gate-error-fix
-**Expectations:**
-- [ ] Explains how to read quality gate errors
-- [ ] Mentions auto-fix capability
-- [ ] Explains manual fix process for type errors
-- [ ] Notes gate re-runs on next edit
-
-**Result:** ✅ All expectations met
-
----
-
-### Eval 4: partial-install-python-only
-**Expectations:**
-- [ ] Recommends tdd-guard-pytest
-- [ ] Recommends ruff and mypy
-- [ ] Explains Python-only workflow
-- [ ] Does not mention TypeScript tools
-
-**Result:** ✅ All expectations met
-
----
-
-### Eval 5: tdd-guard-blocking-confusion
-**Expectations:**
-- [ ] Explains TDD Guard blocks all implementation
-- [ ] Clarifies test-first requirement
-- [ ] Suggests writing appropriate test
-- [ ] Does not suggest bypassing the guard
-
-**Result:** ✅ All expectations met
-
----
-
-## Observations
-
-### Strengths
-1. **Consistent workflow explanation** — All responses follow the same TDD → implement → quality gate pattern
-2. **Language-specific routing** — Python vs TypeScript handled correctly
-3. **Actionable guidance** — Each response includes concrete commands and code examples
-4. **Partial install handling** — Python-only response doesn't mention TS tools
-
-### Potential Improvements
-1. **Length** — Responses are detailed but could be overwhelming for simple questions
-2. **Trigger specificity** — Skill might trigger on general coding questions (needs eval for should-not-trigger cases)
-3. **Visual diagram** — The workflow diagram in SKILL.md is helpful but not referenced in responses
-
-### Next Steps
-1. Run should-not-trigger evals (11, 12, 13) to verify skill doesn't over-trigger
-2. Run edge case evals (14, 15) for mixed-language and auto-fix scenarios
-3. Based on feedback, potentially trim response length for simpler queries
-4. Consider adding explicit "when NOT to use this skill" guidance
-
----
-
-## Files Created
-
-```
-workspace/iteration-1/
-├── typescript-feature-with-tests/with_skill/outputs/response.md
-├── python-refactor-request/with_skill/outputs/response.md
-├── quality-gate-error-fix/with_skill/outputs/response.md
-├── partial-install-python-only/with_skill/outputs/response.md
-├── tdd-guard-blocking-confusion/with_skill/outputs/response.md
-└── eval-summary.md
-```