Metadata-Version: 2.5
Name: custovox-ticket-scorer
Version: 1.0.0
Summary: Support ticket scoring MCP: priority + category + sentiment + escalation risk. Multilingual sentiment. By custovox.ai
Project-URL: Homepage, https://custovox.ai
Project-URL: Repository, https://github.com/jeremdecustovox/custovox-ai
Project-URL: Documentation, https://custovox.ai
Author-email: Jérémie Muselet <jeremie@custovox.ai>
License: MIT
Keywords: customer-support,custovox,mcp,nlp,ticket-scoring,zero-shot
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3.10
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Requires-Python: >=3.10
Requires-Dist: mcp<2,>=1.0.0
Requires-Dist: torch>=2.0.0
Requires-Dist: transformers>=4.40.0
Description-Content-Type: text/markdown

# custovox-ticket-scorer

<!-- mcp-name: io.github.jeremdecustovox/custovox-ticket-scorer -->

Support ticket scoring MCP server.

## What it returns
- **Priority**: urgent / normal / low priority
- **Category**: billing issue, technical problem, delivery issue, product defect, account access, refund request, general inquiry, feature request
- **Sentiment**: Very Negative → Very Positive, multilingual (EN/FR/DE/ES/PT-BR)
- **Escalation risk**: low / medium / high (negative sentiment + urgent priority)

Priority and category use an English zero-shot model (facebook/bart-large-mnli): they are most
accurate on English tickets. Sentiment uses a multilingual model.
Batch mode: up to 20 tickets per call, sorted by escalation risk.

## Installation
pip install custovox-ticket-scorer

The models (about 2 GB) are downloaded from Hugging Face on first start and run on your machine.

## Need more?
**custovox.ai** — Full customer intelligence platform. From €49/month.
https://custovox.ai
