Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    data extraction cleaning structuring pipeline

    1

    Automate the extraction, cleaning, and structuring of messy datasets and text into database-ready tables.

    $19

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    data extraction cleaning structuring pipeline

    data extraction cleaning structuring pipeline

    Example session with this skill installed

    I have a messy CSV named 'leads_raw.csv'. I need to clean it, remove duplicates, and structure it for a Salesforce import. The final file should be 'leads_clean.csv'.

    • Read your context and instructions
    • Compiled the data extraction cleaning
    • Generated the data export

    I have processed 'leads_raw.csv'. I standardized the phone formats, handled 14 missing email entries by flagging them, and removed 5 duplicate records.

    Data Processing Pipeline Result

    FieldStrategy AppliedStatus
    Duplicate RemovalExact match on email and full_name5 rows removed
    FormattingStandardized phone to E.164 and zip_code to 5-digitApplied
    Missing DataFlagged rows with missing email in validation_error column14 rows flagged
    StructureMapped to Salesforce Lead Import Header formatComplete
    ExportSaved to leads_clean.csvSuccess

    Processing Summary

    • Input Rows: 150
    • Cleaned Rows: 131
    • Flagged Rows: 14 (Validation: Missing Email)
    • Deleted Rows: 5 (Duplicates)
    import pandas as pd
    import numpy as np
    
    def clean_leads(input_path, output_path):
        try:
            # Phase 1: Extract
            df = pd.read_csv(input_path)
            
            # Phase 2: Clean
            df.drop_duplicates(subset=['email', 'full_name'], keep='first', inplace=True)
            df['phone'] = df['phone'].astype(str).str.replace(r'\D', '', regex=True)
            
            # Phase 3: Structure & Validate
            df['validation_error'] = np.where(df['email'].isna(), 'Missing Email', '')
            
            # Phase 5: Export
            df.to_csv(output_path, index=False)
            return "Success"
        except Exception as e:
            return f"Error: {str(e)}"
    
    clean_leads('leads_raw.csv', 'leads_clean.csv')
    

    Next steps

    1. Review the validation_error column in leads_clean.csv for the 14 flagged rows.
    2. Upload the leads_clean.csv file directly to the Salesforce Data Import Wizard.
    3. Verify that the phone number formatting matches your specific region requirements.

    leads_raw.csv

    CSV · data export

    Generated

    Example file from a real run - the skill writes it into your workspace.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Messy datasets and unstructured notes stall analysis and break downstream integrations. This skill automates the extraction, cleaning, and structuring of raw data into production-ready formats.

    What it does

    • Source extraction reads data from CSV, Excel, APIs, or raw text notes using pandas.
    • Data sanitization removes duplicates, fixes inconsistent formatting, and standardizes field values.
    • Missing value handling applies configurable fill, drop, or flag strategies based on your requirements.
    • Field extraction parses unstructured text to identify specific entities like dates, names, and amounts.
    • Validation logging appends a reason column to flag specific rows that fail data integrity checks.

    How it works

    1. Define Source Provide the input file or text and specify the desired destination.
    2. Configure Pipeline Select the cleaning strategy for missing data and the list of fields to extract.
    3. Run Transformation The agent executes the extract-clean-structure sequence with row-level error handling.
    4. Review Summary Inspect the processing summary and the exported file ready for CRM or database import.

    Frameworks & tools

    The pipeline is built on Python and pandas for high-performance data manipulation, utilizing try-except blocks for fault tolerance and row-level logging for auditability.

    Why this beats prompting it yourself

    Generic prompts often fail at edge cases like inconsistent date formats or partial null values. This skill enforces a multi-phase engineering pipeline with explicit validation steps, ensuring your output is actually importable into tools like Airtable or Notion.

    Use cases

    • Converting disorganized meeting notes into a structured Airtable-ready table.
    • Normalizing messy customer CSV exports for CRM migration.
    • Building a recurring pipeline to clean weekly API data drops.
    • Extracting financial metrics from raw text logs into structured fields.

    Known limitations

    Requires Python environment with pandas installed. Does not support real-time streaming data or complex multi-table relational joining.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Data & Databases.

    Requires Python and pandas. Best suited for batch processing rather than real-time streams. No support for complex relational joins across multiple tables.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before buying a skill.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 6 days ago

    Accounts
    Google DriveNotionAirtablen8n
    Software
    Required (to use any skill at all):
    Claude Code
    orCursor
    orGitHub Copilot
    Python 3.xPowerShell
    Optional
    , and only for specific skills (not a purchase requirement):

    A host app that supports the Agent Skills format. That's it. copying the skills/ folder is the entire install. only to act on design-unified-google-drive-notion-workspace-two-domain-structure or google-workspace-automation-local-cloud-integration-architecture. only to act on…

    Permissions

    No special permissions declared or detected

    Listed6 days ago

    What's inside

    Frequently Asked Questions

    This skill is designed to handle CSV and Excel files, as well as raw text notes and API data streams. It is particularly effective at turning disorganized text or inconsistent spreadsheets into standardized tables ready for tools like Notion, Airtable, or SQL databases.

    Unlike simple prompts that can be inconsistent, this skill uses a multi-phase engineering pipeline with specific validation logging. It actively handles edge cases like inconsistent date formats and partial null values that typically cause LLMs to hallucinate or break formatting.

    You will receive the complete agent skill configuration, including the pre-built Python/Pandas logic, data sanitization protocols, and the validation logging framework. It includes everything you need to start the extraction and cleaning process immediately within your compatible agent environment.

    This skill requires a Python-capable environment with the Pandas library installed to perform high-performance data manipulation. It is built to be compatible with AI agents that support custom code execution or Python-based tool integration.

    Yes, the pipeline allows you to choose your own cleaning strategy for missing data, such as filling, dropping, or flagging values. You can also define the specific fields you want to extract from unstructured text to match your destination database schema.

    New in Data & Databases