Sentiment Analysis Skill for Auditing Product Feedback Labels
Use Claude Code to organize sentiment classifications by source ID and flag nuanced feedback for review.
中文Install in Claude Code
Use synthetic product comments, an agreed label set, and your project's text classifier. Define what counts as neutral before scoring.
- Get the complete `plugins/ai-ml/sentiment-analysis-tool/skills/analyzing-text-sentiment` directory from the pinned GitHub commit.
- Copy it into the target project's `.claude/skills/`. Check `.claude/skills/analyzing-text-sentiment/SKILL.md`, start Claude Code at the project root, and call `/analyzing-text-sentiment`.
Example task
Classify four comments about secure packaging, shipment as described, a chipped mug, and sarcastic praise for a missing item. Flag the sarcastic record for human review.What to check
The audit includes all four IDs, predicted labels, and per-class counts. Explicitly distinguish direct negatives from possible sarcasm.
Original SKILL.md on GitHub · Updated
Frequently asked question
Can an overall sentiment score replace sampling?
Keep the overall view and sample by class, with extra attention to phrases near the label boundaries.