Text Classifier Pro

Problem addressed

Manual text labeling often fails because results are hard to reuse: inconsistent labels, missing confidence, and unstable CSV column mapping create extra cleanup work. Text Classifier Pro targets this data-organization scenario by turning pasted text, TXT, or CSV input into structured labels plus confidence scores for spreadsheets, reports, or downstream data workflows.

How it works

The core path is: upload text/CSV → AI classification → structured labels + confidence. Common capabilities include:
- Multiple input modes: supports pasted text, TXT, and CSV uploads; CSV can auto-detect the text column and handle GBK, GB2312, and UTF-8.
- Preset and custom templates: includes intent, sentiment, industry, risk, priority, and content classification templates, and supports custom label systems and Prompt configuration.
- Multiple export formats: results can be viewed as a Markdown table and exported to CSV, Excel, or JSON for analysis pipelines.
- Batch and validation behavior: batch processing is rate-limited to about 5 req/sec; API-related features rely on token verification and degrade to a basic mode on network errors instead of blocking use.

Boundaries and notes

This skill fits batch classification tasks with clear label dimensions, such as support intent, sentiment, document industry, or risk signals. If the taxonomy is vague, define classification standards first because output will depend heavily on the Prompt. Note that FREE, STD, PRO, and MAX differ in batch size, history retention, and API access; for complex CSV files, clean column names and text fields first. Inject the API Key via the OPENAI_API_KEY environment variable rather than storing secrets in result files.