菜单
Extract and analyze Excel workbooks (1MB-50MB+) with minimal token usage through local extraction, sheet-based chunking, and persistent caching. Preserves formulas, cell formatting, and complex table structures for compliance matrices, financial models, and large datasets.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Use when applying the same security fix across many files or repos at once — large-scale vulnerability remediation. Provides 7 copy-paste /batch recipes (XSS, HTTPS, SQLi, input validation, log PII, secret rotation, CVE patches) each with a safety tier and false-positive trap, plus scan_diff.py to prove the migration introduced no new HIGH/CRITICAL findings.
Use this skill to install ready-made Claude Code hooks for security — a ConfigChange compliance audit trail (logs every settings/skill change), plus PreToolUse guards and optional HTTP notifications for security-relevant events. Covers command, http, and prompt hook types.
Use this skill to seed Claude Code's auto-memory with pre-built security/compliance patterns — known-safe code patterns, false-positive suppressions for the plugin-security-checker, org compliance rules, and supply-chain best practices. Helps Claude apply consistent security judgment across sessions.
Use this skill to reduce token consumption by 70-80% when using the Chrome DevTools MCP. Applies a snapshot-first strategy (text over images) with decision trees and pattern guides for navigation, forms, debugging, and visual checks, plus optional Gemini Flash vision processing for the cases that genuinely need an image.
Navigate Claude Code documentation and get quick answers about configuration, MCP, plugins, and best practices.
Use this skill to extract and analyze large Word documents (1MB-50MB+) with minimal token usage. Losslessly extracts text, tables, formatting, and structure while achieving 10-50x token reduction through local extraction, semantic chunking by headings, and intelligent caching. Use when a .docx is too large for the context window (policies, contracts, technical reports).
| name | xlsx-smart-extractor |
| description | Extract and analyze Excel workbooks (1MB-50MB+) with minimal token usage through local extraction, sheet-based chunking, and persistent caching. Preserves formulas, cell formatting, and complex table structures for compliance matrices, financial models, and large datasets. |
| license | MIT |
| compatibility | claude-desktop |
| metadata | {"version":"2.2.0","author":"Diego Consolini","category":"productivity","runtime":"python3","dependencies":"requirements.txt"} |
Use this skill when:
⚠️ IMPORTANT: Cache Location
Extracted content is stored in a user cache directory, NOT the working directory:
Cache locations by platform:
~/.claude-cache/xlsx/{workbook_name}_{hash}/C:\Users\{username}\.claude-cache\xlsx\{workbook_name}_{hash}\Why cache directory?
Cache contents:
full_workbook.json - Complete workbook datasheet_{name}.json - Individual sheet filesnamed_ranges.json - Named ranges and tablesmetadata.json - Workbook metadatamanifest.json - Cache manifestAccessing cached content:
# List all cached workbooks
python scripts/query_xlsx.py list
# Query cached content
python scripts/query_xlsx.py search {cache_key} "search query"
# Find cache location (shown in extraction output)
# Example: ~/.claude-cache/xlsx/ComplianceMatrix_a1b2c3d4/
To extract files to working directory:
# Option 1: Use --output-dir flag during extraction
python3 scripts/extract_xlsx.py workbook.xlsx --output-dir ./extracted
# Option 2: Copy from cache manually
cp -r ~/.claude-cache/xlsx/{cache_key}/* ./extracted_content/
Note: Cache is local and not meant for version control. Keep original Excel files in the repository and extract locally on each development machine (one-time operation).
# Extract to cache (default)
python3 scripts/extract_xlsx.py /path/to/workbook.xlsx
# Extract and copy to working directory (interactive prompt)
python3 scripts/extract_xlsx.py /path/to/workbook.xlsx
# Will prompt: "Copy files? (y/n)"
# Will ask: "Keep cache? (y/n)"
# Extract and copy to specific directory (no prompts)
python3 scripts/extract_xlsx.py /path/to/workbook.xlsx --output-dir ./extracted
What happens:
ComplianceMatrix_a8f9e2c1)Output files:
full_workbook.json - All sheets with full datasheets/*.json - Individual sheet data filesformulas.json - All formulas extractedmetadata.json - Workbook metadatanamed_ranges.json - Named ranges and tablesmanifest.json - Extraction summaryPerformance:
python3 scripts/chunk_sheets.py <cache_key>
What happens:
Output files:
chunks/index.json - Chunk metadata and locationschunks/chunk_001.json - Individual chunk datachunks/chunk_002.json - ...Statistics:
# Search by keyword
python3 scripts/query_xlsx.py search <cache_key> "password policy"
# Get specific sheet
python3 scripts/query_xlsx.py sheet <cache_key> "Controls"
# Get cell range
python3 scripts/query_xlsx.py range <cache_key> "Sheet1!A1:E10"
# Get workbook summary
python3 scripts/query_xlsx.py summary <cache_key>
What happens:
Results format:
Found 3 result(s) for query: "password policy"
1. Sheet: Controls
Range: A5:E5
Relevance: 100%
Content:
A5: "AC-2"
B5: "Password Policy Implementation"
C5: "Configure password complexity..."
D5: "Evidence.docx"
E5: "Complete"
Tokens: 85
2. Sheet: Evidence
Range: B12:C12
Relevance: 95%
Content:
B12: "Password policy documented"
C12: "2025-10-15"
Tokens: 32
Total tokens: 117 (vs 45,892 full workbook = 392x reduction)
Scenario: ISO 27001 compliance tracking spreadsheet (5MB, 12 sheets, 500 controls)
Workflow:
python3 scripts/extract_xlsx.py iso27001_controls.xlsxpython3 scripts/query_xlsx.py search iso27001_controls_a8f9e2 "A.9.2.1"python3 scripts/query_xlsx.py sheet iso27001_controls_a8f9e2 "Evidence"Benefits:
Scenario: Revenue projection model (15MB, 8 sheets, complex formulas)
Workflow:
python3 scripts/extract_xlsx.py revenue_model.xlsxpython3 scripts/query_xlsx.py summary revenue_model_f3a8c1python3 scripts/query_xlsx.py search revenue_model_f3a8c1 "formula:SUM"python3 scripts/query_xlsx.py range revenue_model_f3a8c1 "Projections!A1:Z50"Benefits:
Scenario: Security event export (20MB, 50,000 rows, 30 columns)
Workflow:
python3 scripts/extract_xlsx.py security_logs.xlsxpython3 scripts/query_xlsx.py summary security_logs_b9d2e1python3 scripts/query_xlsx.py search security_logs_b9d2e1 "failed"python3 scripts/query_xlsx.py range security_logs_b9d2e1 "Logs!A1:F1000"Benefits:
Not supported:
xlrd separately)pyxlsb separately)odfpy separately)Cell Values:
Cell Formatting:
Formulas:
Workbook Structure:
Small sheets (< 1000 cells):
Wide tables (> 20 columns):
Long tables (> 500 rows):
Named ranges:
Tokens estimated using character count / 4 (approximation):
Actual token usage may vary with model (Claude uses different tokenizer than GPT).
1. File Not Found
Error: Excel file not found: /path/to/file.xlsx
Solution: Verify file path and permissions.
2. Corrupted Workbook
Error: Failed to open workbook: zipfile.BadZipFile
Solution: File may be corrupted. Try opening in Excel and re-saving.
3. Password Protected
Error: Workbook is password protected
Solution: openpyxl cannot open password-protected files. Remove protection first.
4. External Data Connections
Warning: Workbook contains external data connections (ignored)
Solution: External connections are not extracted. Only static data is preserved.
5. Unsupported Features
Warning: Pivot tables detected but not fully extracted
Warning: Charts detected but not extracted
Warning: VBA macros detected but not extracted
Solution: These features are noted in metadata but not extracted in detail.
Large workbook handling:
~/.claude-cache/xlsx/# Python 3.8+
python3 --version
# Install dependencies
pip3 install -r requirements.txt
# Test openpyxl
python3 -c "import openpyxl; print('openpyxl available')"
# Test pandas
python3 -c "import pandas; print('pandas available')"
Solution:
pip3 install openpyxl
Solution:
chmod 755 ~/.claude-cache/xlsx/
Possible causes:
Solution: Use --force flag and check for warnings.
Solution: Process sheets one at a time instead of loading entire workbook.