Analyze Excel chat data to classify conversations as resolved or unresolved.
npx skills add https://github.com/channel-io/sop-agent --skill bot-analysis
Please help me install this Agent Skill. Skill: bot-analysis Source: https://github.com/channel-io/sop-agent/tree/main/.agents/skills/bot-analysis Command: npx skills add https://github.com/channel-io/sop-agent --skill bot-analysis
This Skill analyzes the bot's performance in customer support chats by preprocessing Excel chat data with Python to classify conversations as resolved (bot-only) or unresolved (manager-assisted), enabling data-driven improvements.
Run the analysis pipeline on your Excel chat data to generate the summary and report.
High-intent search queries and answers about installing and using this skill.
Customer-support bot performance is analyzed by preprocessing Excel chat data with Python to classify conversations as resolved or unresolved. The pipeline outputs a summary JSON file and a Markdown report detailing topic-level insights and overall bot metrics.
Classifying bot chat conversations as resolved or unresolved is best handled by running a Python preprocessing script that labels bot-only interactions as resolved and manager-assisted handoffs as unresolved. This automated classification enables data-driven improvements for your support bots.
Yes, bot performance analysis works with topic clustering results. If clustering data exists in your dataset, the analysis pipeline maps conversations to specific topics, enabling you to generate per-topic performance insights and identify patterns in bot behavior within the final report.
You can use Python scripts to preprocess Excel data for customer support topic analysis. The pipeline parses raw chat interactions, identifies bot participation, and structures the data required to classify resolutions and generate topic-level performance reports.
Automating bot-performance analysis yields a bot_analysis_summary.json containing raw metrics and a bot_performance_report.md featuring an executive summary, topic-level patterns, and bot behavior analysis. These outputs provide a comprehensive view of your chatbot's resolution rates and handoff trends.
No, you do not need existing clustering results to run the base analysis, but the pipeline can utilize them if present. Applying clustering results allows the analysis to map conversations to topics and generate detailed per-topic performance insights for your customer-support bots.
joneqianConvert raw FAQ documents into RAG-ready Markdown and Excel files with keywords.
modus-biCreate BI analytics prompts specifying actions, data sources, metrics, and time periods.
EskyeeRank narratives by velocity across Reddit, Twitter, Hacker News, Polymarket, and the open web.
doanchienthangdevProcess unbounded data streams with Kafka Streams, Apache Flink, and ksqlDB.
