Multi-Source Data Cleanser
Paste the following prompt into your AI chat to install this skill:
Please install @user_26bdbd50/multi-source-data-cleaner according to https://skillhub.cn/install/skillhub.md.
About this skill
Problem
Business data often arrives as messy Excel, CSV, JSON, or clipboard text. Field names may not align across systems, order files can contain duplicate rows, phone numbers, dates, amounts, and addresses may be inconsistent, and manual column-by-column cleanup is slow. Merging multiple systems is risky when key fields are missing or conflicting.
How It Works
The skill structures cleaning into composable steps: parse input, identify fields, then clean, fill, format, and merge.
- Multi-format parsing: supports
.xlsx,.xls,CSV,TSV, semi-structuredJSON, and pasted text. - Field identification: detects names, phones, emails, addresses, amounts, dates,
SKU, order IDs, ID numbers, etc., with user-overridable field mapping. - Data cleaning: exact and fuzzy deduplication; missing values can be filled with mean, mode, semantic inference, or left blank; phone, date, amount, and address formats can be standardized.
- Multi-source merge: joins files on key fields; when exact keys are unavailable, fuzzy matching is used, and conflicting fields are resolved by source order or latest timestamp.
- Output: exports
Excel/CSV, writes to Feishu Bitable, and generates aMarkdowndata-quality report.
Boundaries
It is better suited to mid-size spreadsheets and semi-structured data than general-purpose database ETL. Fuzzy dedup uses a fixed 88% threshold, which may over-merge similar names or order IDs; semantic fill and AI tagging depend on external models and need manual review. Pro includes multi-source merge, AI classification, quality reports, and Feishu Bitable output; Free and Basic have limits on rows, sources, and cleaning features. If field semantics are complex, validate field mappings before batch cleaning.
Use Cases
- An e-commerce ops analyst cleans order, refund, and SKU spreadsheets into a unified order detail table.
- A CRM admin standardizes customer contact names, phones, emails, and addresses while removing duplicates.
- A finance analyst reconciles bank statements and internal payment tables by date, amount, and order ID.
- A data analyst reshapes multi-system JSON/CSV exports into an analysis-ready spreadsheet.
Best For
- CRM operations owners who maintain customer master data and need unified contact fields with deduplication.
- E-commerce ops staff handling after-sales or order exceptions who need to merge order and SKU tables.
- Finance specialists reconciling bank statements or payments who need matching by date and amount.
- Data analysts preparing report data who need to normalize messy spreadsheets into analysis-ready formats.
Related Skills
Query Douyin creator videos, image/text posts, and short-drama series by sec_user_id or profile URL with paginated list retrieval.
Uses CDP-controlled browser screenshots of Douyin video pages and multimodal AI to extract teaching points into a Markdown report.
Query Douyin hot searches and works with SocialDataX API or MCP tools, supporting keyword, sorting, time range, duration, content type, and pagination.
Fetch the GitHub Trending leaderboard and generate fact-grounded project analysis, trend hypotheses, and directional insights.