diff --git a/.github/workflows/translate-docs.yml b/.github/workflows/translate-docs.yml
deleted file mode 100644
index 42a6ea5..0000000
--- a/.github/workflows/translate-docs.yml
+++ /dev/null
@@ -1,143 +0,0 @@
-name: Translate Documentation to Chinese
-
-on:
-  push:
-    branches:
-      - main
-      - development
-    paths:
-      - 'docs/**/*.md'
-      - '!docs/zh-CN/**'
-      - '!docs/archive/**'
-  workflow_dispatch:
-    inputs:
-      files:
-        description: 'Specific files to translate (comma-separated, or "all")'
-        required: false
-        default: 'changed'
-
-jobs:
-  detect-changes:
-    runs-on: ubuntu-latest
-    outputs:
-      changed-files: ${{ steps.detect.outputs.files }}
-    steps:
-      - name: Checkout repository
-        uses: actions/checkout@v4
-        with:
-          fetch-depth: 2
-
-      - name: Detect changed files
-        id: detect
-        run: |
-          if [ "${{ github.event.inputs.files }}" = "all" ]; then
-            # Translate all docs
-            FILES=$(find docs -name "*.md" -not -path "docs/zh-CN/*" -not -path "docs/archive/*" | tr '\n' ',')
-          elif [ "${{ github.event.inputs.files }}" != "" ] && [ "${{ github.event.inputs.files }}" != "changed" ]; then
-            # Use provided files
-            FILES="${{ github.event.inputs.files }}"
-          else
-            # Detect changed files
-            FILES=$(git diff --name-only HEAD~1 HEAD | grep "^docs/" | grep -v "^docs/zh-CN/" | grep -v "^docs/archive/" | grep "\.md$" | tr '\n' ',')
-          fi
-          
-          # Remove trailing comma
-          FILES=$(echo "$FILES" | sed 's/,$//')
-          
-          echo "files=$FILES" >> $GITHUB_OUTPUT
-          echo "Detected files: $FILES"
-
-  translate:
-    runs-on: ubuntu-latest
-    needs: detect-changes
-    if: needs.detect-changes.outputs.changed-files != ''
-    steps:
-      - name: Checkout repository
-        uses: actions/checkout@v4
-
-      - name: Set up Python
-        uses: actions/setup-python@v5
-        with:
-          python-version: '3.11'
-
-      - name: Install dependencies
-        run: |
-          pip install anthropic
-
-      - name: Translate documents
-        env:
-          ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
-        run: |
-          IFS=',' read -ra FILES <<< "${{ needs.detect-changes.outputs.changed-files }}"
-          for file in "${FILES[@]}"; do
-            if [ -f "$file" ]; then
-              echo "Translating: $file"
-              python scripts/translate_doc.py "$file" --target-lang zh-CN || echo "Failed: $file"
-            fi
-          done
-
-      - name: Check for changes
-        id: git-check
-        run: |
-          git add docs/zh-CN/
-          if git diff --cached --quiet; then
-            echo "changed=false" >> $GITHUB_OUTPUT
-          else
-            echo "changed=true" >> $GITHUB_OUTPUT
-          fi
-
-      - name: Create Pull Request
-        if: steps.git-check.outputs.changed == 'true'
-        uses: peter-evans/create-pull-request@v6
-        with:
-          token: ${{ secrets.GITHUB_TOKEN }}
-          commit-message: "[Auto] Chinese translation update"
-          title: "🌐 [Auto] Chinese Documentation Translation Update"
-          body: |
-            ## 🇨🇳 中文文档翻译更新 / Chinese Documentation Translation Update
-
-            This PR contains automated translations of updated documentation.
-            
-            ### 变更内容 / Changes
-            ${{ needs.detect-changes.outputs.changed-files }}
-
-            ### 审阅指南 / Review Guide
-            - [ ] 技术术语准确 / Technical terms accurate
-            - [ ] 链接正确指向中文版本 / Links point to Chinese versions
-            - [ ] 代码示例保持原样 / Code examples preserved
-            - [ ] 格式正确 / Formatting correct
-
-            ### 如何审阅 / How to Review
-            1. 查看文件列表 / Check the file list
-            2. 阅读中文翻译 / Read the Chinese translation
-            3. 在 PR 中提出修改建议 / Suggest changes in PR
-            4. 确认后批准 / Approve when satisfied
-
-            ### 相关 Issue
-            - #260 - Chinese Translation
-            
-            ---
-            
-            *This PR was auto-generated by GitHub Actions*
-          branch: auto-translate-zh-cn-${{ github.run_number }}
-          delete-branch: true
-          labels: translation, zh-CN, needs-review, automated
-
-      - name: Update Issue #260
-        if: steps.git-check.outputs.changed == 'true'
-        uses: actions/github-script@v7
-        with:
-          script: |
-            github.rest.issues.createComment({
-              issue_number: 260,
-              owner: context.repo.owner,
-              repo: context.repo.repo,
-              body: `🤖 **自动翻译更新 / Automated Translation Update**
-
-            新的中文翻译已准备就绪，需要社区审阅：
-            - PR: #${{ steps.create-pr.outputs.pull-request-number }}
-            - 文件: ${{ needs.detect-changes.outputs.changed-files }}
-
-            请志愿者帮忙审阅，谢谢！
-            / Community review needed, thanks!`
-            })
diff --git a/=0.24.0 b/=0.24.0
new file mode 100644
index 0000000..83a1f95
--- /dev/null
+++ b/=0.24.0
@@ -0,0 +1,18 @@
+error: externally-managed-environment
+
+× This environment is externally managed
+╰─> To install Python packages system-wide, try 'pacman -S
+    python-xyz', where xyz is the package you are trying to
+    install.
+    
+    If you wish to install a non-Arch-packaged Python package,
+    create a virtual environment using 'python -m venv path/to/venv'.
+    Then use path/to/venv/bin/python and path/to/venv/bin/pip.
+    
+    If you wish to install a non-Arch packaged Python application,
+    it may be easiest to use 'pipx install xyz', which will manage a
+    virtual environment for you. Make sure you have python-pipx
+    installed via pacman.
+
+note: If you believe this is a mistake, please contact your Python installation or OS distribution provider. You can override this, at the risk of breaking your Python installation or OS, by passing --break-system-packages.
+hint: See PEP 668 for the detailed specification.
diff --git a/CHANGELOG.md b/CHANGELOG.md
index 11c4f6c..446f057 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -5,6 +5,18 @@ All notable changes to Skill Seeker 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).
 
+## [Unreleased]
+
+### Changed
+- **Explicit chunk flag names** — All `--chunk-*` flags now include unit suffixes to eliminate ambiguity:
+  - `--chunk-size` (RAG tokens) → `--chunk-tokens`
+  - `--chunk-overlap` (RAG tokens) → `--chunk-overlap-tokens`
+  - `--chunk` (enable RAG chunking) → `--chunk-for-rag`
+  - `--streaming-chunk-size` (chars) → `--streaming-chunk-chars`
+  - `--streaming-overlap` (chars) → `--streaming-overlap-chars`
+  - `--chunk-size` in PDF extractor (pages) → `--pdf-pages-per-chunk`
+- **`setup_logging()` centralized** — Removed duplicate `logging.basicConfig()` calls in `github_scraper.py`, `codebase_scraper.py`, `unified_scraper.py`; all now use shared `setup_logging()` from `utils.py`
+
 ## [3.1.2] - 2026-02-24
 
 ### 🔧 Fix `create` Command Argument Forwarding, Gemini Model, and Enhance Dispatcher
diff --git a/TESTING_GAP_REPORT.md b/TESTING_GAP_REPORT.md
new file mode 100644
index 0000000..2277fb0
--- /dev/null
+++ b/TESTING_GAP_REPORT.md
@@ -0,0 +1,345 @@
+# Comprehensive Testing Gap Report
+
+**Project:** Skill Seekers v3.1.0  
+**Date:** 2026-02-22  
+**Total Test Files:** 113  
+**Total Test Functions:** ~208+ (collected: 2173 tests)
+
+---
+
+## Executive Summary
+
+### Overall Test Health: 🟡 GOOD with Gaps
+
+| Category | Status | Coverage | Key Gaps |
+|----------|--------|----------|----------|
+| CLI Arguments | ✅ Good | 85% | Some edge cases |
+| Workflow System | ✅ Excellent | 90% | Inline stage parsing edge cases |
+| Scrapers | 🟡 Moderate | 70% | Missing real HTTP/PDF tests |
+| Enhancement | 🟡 Partial | 60% | Core logic not tested |
+| MCP Tools | 🟡 Good | 75% | 8 tools not covered |
+| Integration/E2E | 🟡 Moderate | 65% | Heavy mocking |
+| Adaptors | ✅ Good | 80% | Good coverage per platform |
+
+---
+
+## Detailed Findings by Category
+
+### 1. CLI Argument Tests ✅ GOOD
+
+**Files Reviewed:**
+- `test_analyze_command.py` (269 lines, 26 tests)
+- `test_unified.py` - TestUnifiedCLIArguments class (6 tests)
+- `test_pdf_scraper.py` - TestPDFCLIArguments class (4 tests)
+- `test_create_arguments.py` (399 lines)
+- `test_create_integration_basic.py` (310 lines, 23 tests)
+
+**Strengths:**
+- All new workflow flags are tested (`--enhance-workflow`, `--enhance-stage`, `--var`, `--workflow-dry-run`)
+- Argument parsing thoroughly tested
+- Default values verified
+- Complex command combinations tested
+
+**Gaps:**
+- `test_create_integration_basic.py`: 2 tests skipped (source auto-detection not fully tested)
+- No tests for invalid argument combinations beyond basic parsing errors
+
+---
+
+### 2. Workflow Tests ✅ EXCELLENT
+
+**Files Reviewed:**
+- `test_workflow_runner.py` (445 lines, 30+ tests)
+- `test_workflows_command.py` (571 lines, 40+ tests)
+- `test_workflow_tools_mcp.py` (295 lines, 20+ tests)
+
+**Strengths:**
+- Comprehensive workflow execution tests
+- Variable substitution thoroughly tested
+- Dry-run mode tested
+- Workflow chaining tested
+- All 6 workflow subcommands tested (list, show, copy, add, remove, validate)
+- MCP workflow tools tested
+
+**Minor Gaps:**
+- No tests for `_build_inline_engine` edge cases
+- No tests for malformed stage specs (empty, invalid format)
+
+---
+
+### 3. Scraper Tests 🟡 MODERATE with Significant Gaps
+
+**Files Reviewed:**
+- `test_scraper_features.py` (524 lines) - Doc scraper features
+- `test_codebase_scraper.py` (478 lines) - Codebase analysis
+- `test_pdf_scraper.py` (558 lines) - PDF scraper
+- `test_github_scraper.py` (1015 lines) - GitHub scraper
+- `test_unified_analyzer.py` (428 lines) - Unified analyzer
+
+**Critical Gaps:**
+
+#### A. Missing Real External Resource Tests
+| Resource | Test Type | Status |
+|----------|-----------|--------|
+| HTTP Requests (docs) | Mocked only | ❌ Gap |
+| PDF Extraction | Mocked only | ❌ Gap |
+| GitHub API | Mocked only | ❌ Gap (acceptable) |
+| Local Files | Real tests | ✅ Good |
+
+#### B. Missing Core Function Tests
+| Function | Location | Priority |
+|----------|----------|----------|
+| `UnifiedScraper.run()` | unified_scraper.py | 🔴 High |
+| `UnifiedScraper._scrape_documentation()` | unified_scraper.py | 🔴 High |
+| `UnifiedScraper._scrape_github()` | unified_scraper.py | 🔴 High |
+| `UnifiedScraper._scrape_pdf()` | unified_scraper.py | 🔴 High |
+| `UnifiedScraper._scrape_local()` | unified_scraper.py | 🟡 Medium |
+| `DocToSkillConverter.scrape()` | doc_scraper.py | 🔴 High |
+| `PDFToSkillConverter.extract_pdf()` | pdf_scraper.py | 🔴 High |
+
+#### C. PDF Scraper Limited Coverage
+- No actual PDF parsing tests (only mocked)
+- OCR functionality not tested
+- Page range extraction not tested
+
+---
+
+### 4. Enhancement Tests 🟡 PARTIAL - MAJOR GAPS
+
+**Files Reviewed:**
+- `test_enhance_command.py` (367 lines, 25+ tests)
+- `test_enhance_skill_local.py` (163 lines, 14 tests)
+
+**Critical Gap in `test_enhance_skill_local.py`:**
+
+| Function | Lines | Tested? | Priority |
+|----------|-------|---------|----------|
+| `summarize_reference()` | ~50 | ❌ No | 🔴 High |
+| `create_enhancement_prompt()` | ~200 | ❌ No | 🔴 High |
+| `run()` | ~100 | ❌ No | 🔴 High |
+| `_run_headless()` | ~130 | ❌ No | 🔴 High |
+| `_run_background()` | ~80 | ❌ No | 🟡 Medium |
+| `_run_daemon()` | ~60 | ❌ No | 🟡 Medium |
+| `write_status()` | ~30 | ❌ No | 🟡 Medium |
+| `read_status()` | ~40 | ❌ No | 🟡 Medium |
+| `detect_terminal_app()` | ~80 | ❌ No | 🟡 Medium |
+
+**Current Tests Only Cover:**
+- Agent presets configuration
+- Command building
+- Agent name normalization
+- Environment variable handling
+
+**Recommendation:** Add comprehensive tests for the core enhancement logic.
+
+---
+
+### 5. MCP Tool Tests 🟡 GOOD with Coverage Gaps
+
+**Files Reviewed:**
+- `test_mcp_fastmcp.py` (868 lines)
+- `test_mcp_server.py` (715 lines)
+- `test_mcp_vector_dbs.py` (259 lines)
+- `test_real_world_fastmcp.py` (558 lines)
+
+**Coverage Analysis:**
+
+| Tool Category | Tools | Tested | Coverage |
+|---------------|-------|--------|----------|
+| Config Tools | 3 | 3 | ✅ 100% |
+| Scraping Tools | 8 | 4 | 🟡 50% |
+| Packaging Tools | 4 | 4 | ✅ 100% |
+| Splitting Tools | 2 | 2 | ✅ 100% |
+| Source Tools | 5 | 5 | ✅ 100% |
+| Vector DB Tools | 4 | 4 | ✅ 100% |
+| Workflow Tools | 5 | 0 | ❌ 0% |
+| **Total** | **31** | **22** | **🟡 71%** |
+
+**Untested Tools:**
+1. `detect_patterns`
+2. `extract_test_examples`
+3. `build_how_to_guides`
+4. `extract_config_patterns`
+5. `list_workflows`
+6. `get_workflow`
+7. `create_workflow`
+8. `update_workflow`
+9. `delete_workflow`
+
+**Note:** `test_mcp_server.py` tests legacy server, `test_mcp_fastmcp.py` tests modern server.
+
+---
+
+### 6. Integration/E2E Tests 🟡 MODERATE
+
+**Files Reviewed:**
+- `test_create_integration_basic.py` (310 lines)
+- `test_e2e_three_stream_pipeline.py` (598 lines)
+- `test_analyze_e2e.py` (344 lines)
+- `test_install_skill_e2e.py` (533 lines)
+- `test_c3_integration.py` (362 lines)
+
+**Issues Found:**
+
+1. **Skipped Tests:**
+   - `test_create_detects_web_url` - Source auto-detection incomplete
+   - `test_create_invalid_source_shows_error` - Error handling incomplete
+   - `test_cli_via_unified_command` - Asyncio issues
+
+2. **Heavy Mocking:**
+   - Most GitHub API tests use mocking
+   - No real HTTP tests for doc scraping
+   - Integration tests don't test actual integration
+
+3. **Limited Scope:**
+   - Only `--quick` preset tested (not `--comprehensive`)
+   - C3.x tests use mock data only
+   - Most E2E tests are unit tests with mocks
+
+---
+
+### 7. Adaptor Tests ✅ GOOD
+
+**Files Reviewed:**
+- `test_adaptors/test_adaptors_e2e.py` (893 lines)
+- `test_adaptors/test_claude_adaptor.py` (314 lines)
+- `test_adaptors/test_gemini_adaptor.py` (146 lines)
+- `test_adaptors/test_openai_adaptor.py` (188 lines)
+- Plus 8 more platform adaptors
+
+**Strengths:**
+- Each adaptor has dedicated tests
+- Package format testing
+- Upload success/failure scenarios
+- Platform-specific features tested
+
+**Minor Gaps:**
+- Some adaptors only test 1-2 scenarios
+- Error handling coverage varies by platform
+
+---
+
+### 8. Config/Validation Tests ✅ GOOD
+
+**Files Reviewed:**
+- `test_config_validation.py` (270 lines)
+- `test_config_extractor.py` (629 lines)
+- `test_config_fetcher.py` (340 lines)
+
+**Strengths:**
+- Unified vs legacy format detection
+- Field validation comprehensive
+- Error message quality tested
+
+---
+
+## Summary of Critical Testing Gaps
+
+### 🔴 HIGH PRIORITY (Must Fix)
+
+1. **Enhancement Core Logic**
+   - File: `test_enhance_skill_local.py`
+   - Missing: 9 major functions
+   - Impact: Core feature untested
+
+2. **Unified Scraper Main Flow**
+   - File: New tests needed
+   - Missing: `_scrape_*()` methods, `run()` orchestration
+   - Impact: Multi-source scraping untested
+
+3. **Actual HTTP/PDF/GitHub Integration**
+   - Missing: Real external resource tests
+   - Impact: Only mock tests exist
+
+### 🟡 MEDIUM PRIORITY (Should Fix)
+
+4. **MCP Workflow Tools**
+   - Missing: 5 workflow tools (0% coverage)
+   - Impact: MCP workflow features untested
+
+5. **Skipped Integration Tests**
+   - 3 tests skipped
+   - Impact: Source auto-detection incomplete
+
+6. **PDF Real Extraction**
+   - Missing: Actual PDF parsing
+   - Impact: PDF feature quality unknown
+
+### 🟢 LOW PRIORITY (Nice to Have)
+
+7. **Additional Scraping Tools**
+   - Missing: 4 scraping tool tests
+   - Impact: Low (core tools covered)
+
+8. **Edge Case Coverage**
+   - Missing: Invalid argument combinations
+   - Impact: Low (happy path covered)
+
+---
+
+## Recommendations
+
+### Immediate Actions (Next Sprint)
+
+1. **Add Enhancement Logic Tests** (~400 lines)
+   - Test `summarize_reference()`
+   - Test `create_enhancement_prompt()`
+   - Test `run()` method
+   - Test status read/write
+
+2. **Fix Skipped Tests** (~100 lines)
+   - Fix asyncio issues in `test_cli_via_unified_command`
+   - Complete source auto-detection tests
+
+3. **Add MCP Workflow Tool Tests** (~200 lines)
+   - Test all 5 workflow tools
+
+### Short Term (Next Month)
+
+4. **Add Unified Scraper Integration Tests** (~300 lines)
+   - Test main orchestration flow
+   - Test individual source scraping
+
+5. **Add Real PDF Tests** (~150 lines)
+   - Test with actual PDF files
+   - Test OCR if available
+
+### Long Term (Next Quarter)
+
+6. **HTTP Integration Tests** (~200 lines)
+   - Test with real websites (use test sites)
+   - Mock server approach
+
+7. **Complete E2E Pipeline** (~300 lines)
+   - Full workflow from scrape to upload
+   - Real GitHub repo (fork test repo)
+
+---
+
+## Test Quality Metrics
+
+| Metric | Score | Notes |
+|--------|-------|-------|
+| Test Count | 🟢 Good | 2173+ tests |
+| Coverage | 🟡 Moderate | ~75% estimated |
+| Real Tests | 🟡 Moderate | Many mocked |
+| Documentation | 🟢 Good | Most tests documented |
+| Maintenance | 🟢 Good | Tests recently updated |
+
+---
+
+## Conclusion
+
+The Skill Seekers test suite is **comprehensive in quantity** (2173+ tests) but has **quality gaps** in critical areas:
+
+1. **Core enhancement logic** is largely untested
+2. **Multi-source scraping** orchestration lacks integration tests
+3. **MCP workflow tools** have zero coverage
+4. **Real external resource** testing is minimal
+
+**Priority:** Fix the 🔴 HIGH priority gaps first, as they impact core functionality.
+
+---
+
+*Report generated: 2026-02-22*  
+*Reviewer: Systematic test review with parallel subagent analysis*
diff --git a/docs/archive/legacy/QUICK_REFERENCE.md b/docs/archive/legacy/QUICK_REFERENCE.md
index 7004be7..67ca817 100644
--- a/docs/archive/legacy/QUICK_REFERENCE.md
+++ b/docs/archive/legacy/QUICK_REFERENCE.md
@@ -71,7 +71,7 @@ skill-seekers pdf manual.pdf --name product-manual
 skill-seekers pdf scanned.pdf --enable-ocr
 
 # Large PDF (chunked processing)
-skill-seekers pdf large.pdf --chunk-size 50
+skill-seekers pdf large.pdf --pdf-pages-per-chunk 50
 ```
 
 ### Multi-Source Scraping
diff --git a/docs/archive/research/PDF_IMAGE_EXTRACTION.md b/docs/archive/research/PDF_IMAGE_EXTRACTION.md
index 9d17186..26b58e8 100644
--- a/docs/archive/research/PDF_IMAGE_EXTRACTION.md
+++ b/docs/archive/research/PDF_IMAGE_EXTRACTION.md
@@ -122,7 +122,7 @@ python3 cli/pdf_extractor_poc.py documentation.pdf \
   --extract-images \
   --min-image-size 150 \
   --min-quality 6.0 \
-  --chunk-size 20 \
+  --pdf-pages-per-chunk 20 \
   --output documentation.json \
   --verbose \
   --pretty
@@ -477,7 +477,7 @@ python3 cli/pdf_extractor_poc.py manual.pdf \
   --image-dir assets/images/ \
   --min-image-size 200 \
   --min-quality 7.0 \
-  --chunk-size 15 \
+  --pdf-pages-per-chunk 15 \
   --output manual.json \
   --verbose \
   --pretty
diff --git a/docs/features/PDF_CHUNKING.md b/docs/features/PDF_CHUNKING.md
index 1ff8a48..93ca392 100644
--- a/docs/features/PDF_CHUNKING.md
+++ b/docs/features/PDF_CHUNKING.md
@@ -25,10 +25,10 @@ Break large PDFs into smaller, manageable chunks:
 python3 cli/pdf_extractor_poc.py input.pdf
 
 # Custom chunk size (20 pages per chunk)
-python3 cli/pdf_extractor_poc.py input.pdf --chunk-size 20
+python3 cli/pdf_extractor_poc.py input.pdf --pdf-pages-per-chunk 20
 
 # Disable chunking (single chunk with all pages)
-python3 cli/pdf_extractor_poc.py input.pdf --chunk-size 0
+python3 cli/pdf_extractor_poc.py input.pdf --pdf-pages-per-chunk 0
 ```
 
 ### ✅ 2. Chapter/Section Detection
@@ -272,7 +272,7 @@ cat manual.json | jq '.total_chunks'
 
 ```bash
 # Large PDF with bigger chunks (50 pages each)
-python3 cli/pdf_extractor_poc.py large_manual.pdf --chunk-size 50 -o output.json -v
+python3 cli/pdf_extractor_poc.py large_manual.pdf --pdf-pages-per-chunk 50 -o output.json -v
 
 # Verbose output shows:
 # 📦 Creating chunks (chunk_size=50)...
@@ -286,7 +286,7 @@ python3 cli/pdf_extractor_poc.py large_manual.pdf --chunk-size 50 -o output.json
 
 ```bash
 # Process all pages as single chunk
-python3 cli/pdf_extractor_poc.py small_doc.pdf --chunk-size 0 -o output.json
+python3 cli/pdf_extractor_poc.py small_doc.pdf --pdf-pages-per-chunk 0 -o output.json
 ```
 
 ---
@@ -369,7 +369,7 @@ Create a test PDF with chapters:
 3. Page 30: "Chapter 3: API Reference"
 
 ```bash
-python3 cli/pdf_extractor_poc.py test.pdf -o test.json --chunk-size 20 -v
+python3 cli/pdf_extractor_poc.py test.pdf -o test.json --pdf-pages-per-chunk 20 -v
 
 # Verify chapters detected
 cat test.json | jq '.chapters'
@@ -441,7 +441,7 @@ The chunking feature lays groundwork for:
 **Example workflow:**
 ```bash
 # Extract large manual with chapters
-python3 cli/pdf_extractor_poc.py large_manual.pdf --chunk-size 25 -o manual.json
+python3 cli/pdf_extractor_poc.py large_manual.pdf --pdf-pages-per-chunk 25 -o manual.json
 
 # Future: Build skill from chunks
 python3 cli/build_skill_from_pdf.py manual.json
diff --git a/docs/integrations/CHROMA.md b/docs/integrations/CHROMA.md
index 7dc6238..b016bff 100644
--- a/docs/integrations/CHROMA.md
+++ b/docs/integrations/CHROMA.md
@@ -223,7 +223,7 @@ skill-seekers package output/codebase --target langchain
 
 **Option D: RAG-Optimized Chunking**
 ```bash
-skill-seekers scrape --config configs/fastapi.json --chunk-for-rag --chunk-size 512
+skill-seekers scrape --config configs/fastapi.json --chunk-for-rag --chunk-tokens 512
 skill-seekers package output/fastapi --target langchain
 ```
 
@@ -968,7 +968,7 @@ collection.add(
 
 2. **Implement Semantic Chunking:**
    ```bash
-   skill-seekers scrape --config configs/fastapi.json --chunk-for-rag --chunk-size 512
+   skill-seekers scrape --config configs/fastapi.json --chunk-for-rag --chunk-tokens 512
    ```
 
 3. **Set Up Multi-Collection Search:**
diff --git a/docs/integrations/FAISS.md b/docs/integrations/FAISS.md
index e9324b7..658f0a4 100644
--- a/docs/integrations/FAISS.md
+++ b/docs/integrations/FAISS.md
@@ -255,7 +255,7 @@ skill-seekers package output/codebase --target langchain
 
 **Option D: RAG-Optimized Chunking**
 ```bash
-skill-seekers scrape --config configs/fastapi.json --chunk-for-rag --chunk-size 512
+skill-seekers scrape --config configs/fastapi.json --chunk-for-rag --chunk-tokens 512
 skill-seekers package output/fastapi --target langchain
 ```
 
diff --git a/docs/integrations/HAYSTACK.md b/docs/integrations/HAYSTACK.md
index 47a087c..9f8ddcb 100644
--- a/docs/integrations/HAYSTACK.md
+++ b/docs/integrations/HAYSTACK.md
@@ -318,8 +318,8 @@ print(response["llm"]["replies"][0])
 # Enable semantic chunking (preserves code blocks, respects paragraphs)
 skill-seekers scrape --config configs/django.json \
   --chunk-for-rag \
-  --chunk-size 512 \
-  --chunk-overlap 50
+  --chunk-tokens 512 \
+  --chunk-overlap-tokens 50
 
 # Package chunked output
 skill-seekers package output/django --target haystack
@@ -439,8 +439,8 @@ python scripts/merge_documents.py \
 # Enable chunking for frameworks with long pages
 skill-seekers scrape --config configs/django.json \
   --chunk-for-rag \
-  --chunk-size 512 \
-  --chunk-overlap 50
+  --chunk-tokens 512 \
+  --chunk-overlap-tokens 50
 ```
 
 ### 2. Choose Right Document Store
@@ -506,8 +506,8 @@ Complete example of building a FastAPI documentation chatbot:
 # Scrape FastAPI docs with chunking
 skill-seekers scrape --config configs/fastapi.json \
   --chunk-for-rag \
-  --chunk-size 512 \
-  --chunk-overlap 50 \
+  --chunk-tokens 512 \
+  --chunk-overlap-tokens 50 \
   --max-pages 200
 
 # Package for Haystack
@@ -698,8 +698,8 @@ skill-seekers scrape --config configs/fastapi.json --chunk-for-rag
 # 2. Adjust chunk size
 skill-seekers scrape --config configs/fastapi.json \
   --chunk-for-rag \
-  --chunk-size 768 \  # Larger chunks for more context
-  --chunk-overlap 100  # More overlap for continuity
+  --chunk-tokens 768 \  # Larger chunks for more context
+  --chunk-overlap-tokens 100  # More overlap for continuity
 
 # 3. Use hybrid search (BM25 + embeddings)
 # See Advanced Usage section
diff --git a/docs/integrations/QDRANT.md b/docs/integrations/QDRANT.md
index 9dec5e3..beaf8d7 100644
--- a/docs/integrations/QDRANT.md
+++ b/docs/integrations/QDRANT.md
@@ -270,7 +270,7 @@ skill-seekers package output/codebase --target langchain
 
 **Option D: RAG-Optimized Chunking**
 ```bash
-skill-seekers scrape --config configs/fastapi.json --chunk-for-rag --chunk-size 512
+skill-seekers scrape --config configs/fastapi.json --chunk-for-rag --chunk-tokens 512
 skill-seekers package output/fastapi --target langchain
 ```
 
diff --git a/docs/integrations/WEAVIATE.md b/docs/integrations/WEAVIATE.md
index dd17a0e..1648ba5 100644
--- a/docs/integrations/WEAVIATE.md
+++ b/docs/integrations/WEAVIATE.md
@@ -210,7 +210,7 @@ skill-seekers package output/codebase --target langchain
 
 **Option D: RAG-Optimized Chunking**
 ```bash
-skill-seekers scrape --config configs/fastapi.json --chunk-for-rag --chunk-size 512
+skill-seekers scrape --config configs/fastapi.json --chunk-for-rag --chunk-tokens 512
 skill-seekers package output/fastapi --target langchain
 ```
 
@@ -960,7 +960,7 @@ print(schema.get("multiTenancyConfig", {}).get("enabled"))  # Should be True
 
 2. **Implement Semantic Chunking:**
    ```bash
-   skill-seekers scrape --config configs/fastapi.json --chunk-for-rag --chunk-size 512
+   skill-seekers scrape --config configs/fastapi.json --chunk-for-rag --chunk-tokens 512
    ```
 
 3. **Set Up Multi-Tenancy:**
diff --git a/docs/reference/CLI_REFERENCE.md b/docs/reference/CLI_REFERENCE.md
index 7fac99b..f5be01f 100644
--- a/docs/reference/CLI_REFERENCE.md
+++ b/docs/reference/CLI_REFERENCE.md
@@ -252,8 +252,8 @@ skill-seekers create [source] [options]
 | | `--workflow-dry-run` | | Preview workflow without executing |
 | | `--dry-run` | | Preview without creating |
 | | `--chunk-for-rag` | | Enable RAG chunking |
-| | `--chunk-size` | 512 | Chunk size in tokens |
-| | `--chunk-overlap` | 50 | Chunk overlap in tokens |
+| | `--chunk-tokens` | 512 | Chunk size in tokens |
+| | `--chunk-overlap-tokens` | 50 | Chunk overlap in tokens |
 | | `--help-web` | | Show web scraping options |
 | | `--help-github` | | Show GitHub options |
 | | `--help-local` | | Show local analysis options |
@@ -615,10 +615,10 @@ skill-seekers package SKILL_DIRECTORY [options]
 | | `--skip-quality-check` | | Skip quality checks |
 | | `--upload` | | Auto-upload after packaging |
 | | `--streaming` | | Streaming mode for large docs |
-| | `--chunk-size` | 4000 | Max chars per chunk (streaming) |
-| | `--chunk-overlap` | 200 | Overlap between chunks |
+| | `--streaming-chunk-chars` | 4000 | Max chars per chunk (streaming) |
+| | `--streaming-overlap-chars` | 200 | Overlap between chunks (chars) |
 | | `--batch-size` | 100 | Chunks per batch |
-| | `--chunk` | | Enable RAG chunking |
+| | `--chunk-for-rag` | | Enable RAG chunking |
 | | `--chunk-tokens` | 512 | Max tokens per chunk |
 | | `--no-preserve-code` | | Allow code block splitting |
 
@@ -877,7 +877,7 @@ skill-seekers stream --config CONFIG [options]
 | Short | Long | Description |
 |-------|------|-------------|
 | `-c` | `--config` | Config JSON file |
-| | `--chunk-size` | Size of each chunk |
+| | `--streaming-chunk-chars` | Maximum characters per chunk (default: 4000) |
 | | `--output` | Output directory |
 
 **Examples:**
@@ -887,7 +887,7 @@ skill-seekers stream --config CONFIG [options]
 skill-seekers stream --config configs/large-docs.json
 
 # Custom chunk size
-skill-seekers stream --config configs/large-docs.json --chunk-size 1000
+skill-seekers stream --config configs/large-docs.json --streaming-chunk-chars 1000
 ```
 
 ---
diff --git a/docs/strategy/ACTION_PLAN.md b/docs/strategy/ACTION_PLAN.md
index 1be4d2f..e49a903 100644
--- a/docs/strategy/ACTION_PLAN.md
+++ b/docs/strategy/ACTION_PLAN.md
@@ -365,8 +365,8 @@ Position Skill Seekers as **the universal documentation preprocessor** for the e
 2. **Implement Chunking for RAG** (8-12 hours)
    ```bash
    skill-seekers scrape --chunk-for-rag \
-       --chunk-size 512 \
-       --chunk-overlap 50 \
+       --chunk-tokens 512 \
+       --chunk-overlap-tokens 50 \
        --preserve-code-blocks
    ```
 
diff --git a/docs/strategy/KIMI_ANALYSIS_COMPARISON.md b/docs/strategy/KIMI_ANALYSIS_COMPARISON.md
index 5d43714..2cefce8 100644
--- a/docs/strategy/KIMI_ANALYSIS_COMPARISON.md
+++ b/docs/strategy/KIMI_ANALYSIS_COMPARISON.md
@@ -139,8 +139,8 @@ skill-seekers scrape --format confluence     # Confluence storage format
 ```bash
 # New flag for embedding-optimized chunking
 skill-seekers scrape --chunk-for-rag \
-    --chunk-size 512 \
-    --chunk-overlap 50 \
+    --chunk-tokens 512 \
+    --chunk-overlap-tokens 50 \
     --add-metadata
 
 # Output: chunks with metadata for embedding
diff --git a/docs/user-guide/02-scraping.md b/docs/user-guide/02-scraping.md
index d54a016..63e448a 100644
--- a/docs/user-guide/02-scraping.md
+++ b/docs/user-guide/02-scraping.md
@@ -385,7 +385,7 @@ skill-seekers create <url> --max-pages 100
 skill-seekers create <url> --streaming
 
 # Or smaller chunks
-skill-seekers create <url> --chunk-size 500
+skill-seekers create <url> --chunk-tokens 500
 ```
 
 ---
diff --git a/docs/user-guide/04-packaging.md b/docs/user-guide/04-packaging.md
index 847453c..cced71a 100644
--- a/docs/user-guide/04-packaging.md
+++ b/docs/user-guide/04-packaging.md
@@ -158,8 +158,8 @@ skill-seekers package output/large-skill/ --streaming
 # Custom chunk size
 skill-seekers package output/large-skill/ \
   --streaming \
-  --chunk-size 2000 \
-  --chunk-overlap 100
+  --streaming-chunk-chars 2000 \
+  --streaming-overlap-chars 100
 ```
 
 **When to use:**
@@ -177,23 +177,23 @@ Optimize for Retrieval-Augmented Generation:
 # Enable semantic chunking
 skill-seekers package output/my-skill/ \
   --target langchain \
-  --chunk \
+  --chunk-for-rag \
   --chunk-tokens 512
 
 # Custom chunk size
 skill-seekers package output/my-skill/ \
   --target chroma \
   --chunk-tokens 256 \
-  --chunk-overlap 50
+  --chunk-overlap-tokens 50
 ```
 
 **Chunking Options:**
 
 | Option | Default | Description |
 |--------|---------|-------------|
-| `--chunk` | auto | Enable chunking |
+| `--chunk-for-rag` | auto | Enable chunking |
 | `--chunk-tokens` | 512 | Tokens per chunk |
-| `--chunk-overlap` | 50 | Overlap between chunks |
+| `--chunk-overlap-tokens` | 50 | Overlap between chunks (tokens) |
 | `--no-preserve-code` | - | Allow splitting code blocks |
 
 ---
@@ -449,7 +449,7 @@ skill-seekers upload output/my-skill-claude.zip --target claude
 skill-seekers package output/my-skill/ --streaming
 
 # Smaller chunks
-skill-seekers package output/my-skill/ --streaming --chunk-size 1000
+skill-seekers package output/my-skill/ --streaming --streaming-chunk-chars 1000
 ```
 
 ---
diff --git a/docs/user-guide/06-troubleshooting.md b/docs/user-guide/06-troubleshooting.md
index 00d01b0..88042f3 100644
--- a/docs/user-guide/06-troubleshooting.md
+++ b/docs/user-guide/06-troubleshooting.md
@@ -295,7 +295,7 @@ skill-seekers package output/my-skill/ --streaming
 # Reduce chunk size
 skill-seekers package output/my-skill/ \
   --streaming \
-  --chunk-size 1000
+  --streaming-chunk-chars 1000
 ```
 
 ---
diff --git a/docs/zh-CN/reference/CLI_REFERENCE.md b/docs/zh-CN/reference/CLI_REFERENCE.md
index 5b64ba8..88ffbc0 100644
--- a/docs/zh-CN/reference/CLI_REFERENCE.md
+++ b/docs/zh-CN/reference/CLI_REFERENCE.md
@@ -237,8 +237,8 @@ skill-seekers create [source] [options]
 | | `--workflow-dry-run` | | Preview workflow without executing |
 | | `--dry-run` | | Preview without creating |
 | | `--chunk-for-rag` | | Enable RAG chunking |
-| | `--chunk-size` | 512 | Chunk size in tokens |
-| | `--chunk-overlap` | 50 | Chunk overlap in tokens |
+| | `--chunk-tokens` | 512 | Chunk size in tokens |
+| | `--chunk-overlap-tokens` | 50 | Chunk overlap in tokens |
 | | `--help-web` | | Show web scraping options |
 | | `--help-github` | | Show GitHub options |
 | | `--help-local` | | Show local analysis options |
@@ -593,10 +593,10 @@ skill-seekers package SKILL_DIRECTORY [options]
 | | `--skip-quality-check` | | Skip quality checks |
 | | `--upload` | | Auto-upload after packaging |
 | | `--streaming` | | Streaming mode for large docs |
-| | `--chunk-size` | 4000 | Max chars per chunk (streaming) |
-| | `--chunk-overlap` | 200 | Overlap between chunks |
+| | `--streaming-chunk-chars` | 4000 | Max chars per chunk (streaming) |
+| | `--streaming-overlap-chars` | 200 | Overlap between chunks (chars) |
 | | `--batch-size` | 100 | Chunks per batch |
-| | `--chunk` | | Enable RAG chunking |
+| | `--chunk-for-rag` | | Enable RAG chunking |
 | | `--chunk-tokens` | 512 | Max tokens per chunk |
 | | `--no-preserve-code` | | Allow code block splitting |
 
@@ -847,7 +847,7 @@ skill-seekers stream --config CONFIG [options]
 | Short | Long | Description |
 |-------|------|-------------|
 | `-c` | `--config` | Config JSON file |
-| | `--chunk-size` | Size of each chunk |
+| | `--streaming-chunk-chars` | Maximum characters per chunk (default: 4000) |
 | | `--output` | Output directory |
 
 **Examples:**
@@ -857,7 +857,7 @@ skill-seekers stream --config CONFIG [options]
 skill-seekers stream --config configs/large-docs.json
 
 # Custom chunk size
-skill-seekers stream --config configs/large-docs.json --chunk-size 1000
+skill-seekers stream --config configs/large-docs.json --streaming-chunk-chars 1000
 ```
 
 ---
diff --git a/docs/zh-CN/user-guide/02-scraping.md b/docs/zh-CN/user-guide/02-scraping.md
index d54a016..63e448a 100644
--- a/docs/zh-CN/user-guide/02-scraping.md
+++ b/docs/zh-CN/user-guide/02-scraping.md
@@ -385,7 +385,7 @@ skill-seekers create <url> --max-pages 100
 skill-seekers create <url> --streaming
 
 # Or smaller chunks
-skill-seekers create <url> --chunk-size 500
+skill-seekers create <url> --chunk-tokens 500
 ```
 
 ---
diff --git a/docs/zh-CN/user-guide/04-packaging.md b/docs/zh-CN/user-guide/04-packaging.md
index 847453c..cced71a 100644
--- a/docs/zh-CN/user-guide/04-packaging.md
+++ b/docs/zh-CN/user-guide/04-packaging.md
@@ -158,8 +158,8 @@ skill-seekers package output/large-skill/ --streaming
 # Custom chunk size
 skill-seekers package output/large-skill/ \
   --streaming \
-  --chunk-size 2000 \
-  --chunk-overlap 100
+  --streaming-chunk-chars 2000 \
+  --streaming-overlap-chars 100
 ```
 
 **When to use:**
@@ -177,23 +177,23 @@ Optimize for Retrieval-Augmented Generation:
 # Enable semantic chunking
 skill-seekers package output/my-skill/ \
   --target langchain \
-  --chunk \
+  --chunk-for-rag \
   --chunk-tokens 512
 
 # Custom chunk size
 skill-seekers package output/my-skill/ \
   --target chroma \
   --chunk-tokens 256 \
-  --chunk-overlap 50
+  --chunk-overlap-tokens 50
 ```
 
 **Chunking Options:**
 
 | Option | Default | Description |
 |--------|---------|-------------|
-| `--chunk` | auto | Enable chunking |
+| `--chunk-for-rag` | auto | Enable chunking |
 | `--chunk-tokens` | 512 | Tokens per chunk |
-| `--chunk-overlap` | 50 | Overlap between chunks |
+| `--chunk-overlap-tokens` | 50 | Overlap between chunks (tokens) |
 | `--no-preserve-code` | - | Allow splitting code blocks |
 
 ---
@@ -449,7 +449,7 @@ skill-seekers upload output/my-skill-claude.zip --target claude
 skill-seekers package output/my-skill/ --streaming
 
 # Smaller chunks
-skill-seekers package output/my-skill/ --streaming --chunk-size 1000
+skill-seekers package output/my-skill/ --streaming --streaming-chunk-chars 1000
 ```
 
 ---
diff --git a/docs/zh-CN/user-guide/06-troubleshooting.md b/docs/zh-CN/user-guide/06-troubleshooting.md
index 00d01b0..88042f3 100644
--- a/docs/zh-CN/user-guide/06-troubleshooting.md
+++ b/docs/zh-CN/user-guide/06-troubleshooting.md
@@ -295,7 +295,7 @@ skill-seekers package output/my-skill/ --streaming
 # Reduce chunk size
 skill-seekers package output/my-skill/ \
   --streaming \
-  --chunk-size 1000
+  --streaming-chunk-chars 1000
 ```
 
 ---
diff --git a/examples/haystack-pipeline/README.md b/examples/haystack-pipeline/README.md
index d40ba08..070a7a1 100644
--- a/examples/haystack-pipeline/README.md
+++ b/examples/haystack-pipeline/README.md
@@ -132,7 +132,7 @@ For better retrieval quality, use semantic chunking:
 
 ```bash
 # Generate with chunking
-skill-seekers scrape --config configs/react.json --max-pages 100 --chunk-for-rag --chunk-size 512 --chunk-overlap 50
+skill-seekers scrape --config configs/react.json --max-pages 100 --chunk-for-rag --chunk-tokens 512 --chunk-overlap-tokens 50
 
 # Use chunked output
 python quickstart.py --chunked
diff --git a/src/skill_seekers/workflows/comparison-matrix.yaml b/src/skill_seekers/workflows/comparison-matrix.yaml
index 669452f..d87dabe 100644
--- a/src/skill_seekers/workflows/comparison-matrix.yaml
+++ b/src/skill_seekers/workflows/comparison-matrix.yaml
@@ -6,7 +6,6 @@ applies_to:
   - doc_scraping
 variables:
   depth: comprehensive
-  alternatives: []
 stages:
   - name: feature_comparison
     type: custom
diff --git a/src/skill_seekers/workflows/data-validation.yaml b/src/skill_seekers/workflows/data-validation.yaml
index 198ebff..fa31efb 100644
--- a/src/skill_seekers/workflows/data-validation.yaml
+++ b/src/skill_seekers/workflows/data-validation.yaml
@@ -164,5 +164,5 @@ post_process:
   add_metadata:
     enhanced: true
     workflow: data-validation
-    domain: ml
+    domain: backend
     has_validation_docs: true
diff --git a/src/skill_seekers/workflows/default.yaml b/src/skill_seekers/workflows/default.yaml
index f5329d3..7992352 100644
--- a/src/skill_seekers/workflows/default.yaml
+++ b/src/skill_seekers/workflows/default.yaml
@@ -17,6 +17,46 @@ stages:
     target: examples
     enabled: true
     uses_history: false
+  - name: architecture_overview
+    type: custom
+    target: architecture
+    uses_history: false
+    enabled: true
+    prompt: >
+      Provide a concise architectural overview of this codebase.
+
+      Cover:
+      1. Overall architecture style (MVC, microservices, layered, etc.)
+      2. Key components and their responsibilities
+      3. Data flow between components
+      4. External dependencies and integrations
+      5. Entry points (CLI, API, web, etc.)
+
+      Output JSON with:
+      - "architecture_style": main architectural pattern
+      - "components": array of {name, responsibility}
+      - "data_flow": how data moves through the system
+      - "external_deps": third-party services and libraries
+      - "entry_points": how users interact with the system
+  - name: skill_polish
+    type: custom
+    target: skill_md
+    uses_history: true
+    enabled: true
+    prompt: >
+      Review the SKILL.md content generated so far and improve it.
+
+      Fix:
+      1. Unclear or overly technical descriptions
+      2. Missing quick-start examples
+      3. Gaps in the overview section
+      4. Redundant or duplicate information
+      5. Formatting inconsistencies
+
+      Output JSON with:
+      - "improved_overview": rewritten overview section
+      - "quick_start": concise getting-started snippet
+      - "key_concepts": 3-5 essential concepts a developer needs to know
 post_process:
   reorder_sections: []
   add_metadata:
diff --git a/src/skill_seekers/workflows/minimal.yaml b/src/skill_seekers/workflows/minimal.yaml
index a8f4a23..b4f5919 100644
--- a/src/skill_seekers/workflows/minimal.yaml
+++ b/src/skill_seekers/workflows/minimal.yaml
@@ -14,12 +14,17 @@ stages:
     uses_history: false
     enabled: true
     prompt: >
-      Review the following SKILL.md content and make minimal improvements:
-      - Fix obvious formatting issues
-      - Ensure the overview section is clear and concise
-      - Remove duplicate or redundant information
+      Review the SKILL.md content and make minimal targeted improvements.
 
-      Return the improved content as plain text without extra commentary.
+      Fix only:
+      1. Obvious formatting issues (broken lists, inconsistent headers)
+      2. Unclear overview section (make it one clear paragraph)
+      3. Duplicate or redundant information (remove repeats)
+
+      Output JSON with:
+      - "improved_overview": rewritten overview paragraph (plain markdown)
+      - "removed_sections": list of section names that were removed as duplicates
+      - "formatting_fixes": list of specific formatting issues corrected
 post_process:
   reorder_sections: []
   add_metadata:
diff --git a/src/skill_seekers/workflows/security-focus.yaml b/src/skill_seekers/workflows/security-focus.yaml
index c2f7923..c6d2e80 100644
--- a/src/skill_seekers/workflows/security-focus.yaml
+++ b/src/skill_seekers/workflows/security-focus.yaml
@@ -3,9 +3,7 @@ description: "Security-focused review: vulnerabilities, auth, data handling"
 version: "1.0"
 applies_to:
   - codebase_analysis
-  - python
-  - javascript
-  - typescript
+  - github_analysis
 variables:
   depth: comprehensive
 stages:
diff --git a/uv.lock b/uv.lock
index b16a7ef..ca708f7 100644
--- a/uv.lock
+++ b/uv.lock
@@ -5204,7 +5204,7 @@ wheels = [
 
 [[package]]
 name = "skill-seekers"
-version = "3.1.1"
+version = "3.1.2"
 source = { editable = "." }
 dependencies = [
     { name = "anthropic" },