Prepare a first reply to a customer inquiry by classifying the issue, confirming understanding, and requesting missing details. Use existing documentation for answers, acknowledge feature requests, or route cases that need human follow-up.
First-response playbook for an inbound customer message — classifies the ticket, restates the issue, gathers missing context, and routes the reply.
Review the instructions the agent will receive before you install.
A playbook for the first response to a new customer message.
A customer message has arrived and you have not yet replied to it.
bug — something is broken or behaving unexpectedly,question — the customer wants information,feature — a request for new behavior,escalation — out-of-policy, refund > $100, abusive, legal, or
compliance-flavored.bug or question: search workspace docs (grep_files,
memory_search) and prior sessions (session_search) for an existing
answer; reply with the solution and a citation.feature: thank them, log the request in your reply (the runtime
captures it), and set expectations.escalation: hand off to escalation-router and tell the customer
a human will follow up within one business day.AGENTS.md.A single customer-facing message. Do not include the classification label in the message body — keep it as the first line of your internal scratch.
No reviews yet. Tell others what you thought.
No login needed. Share how it worked for you, ask a question, or note where it did not run.
No comments yet. Be the first to write one.
Build Java applications with cooperating AI agents using AgentScope Java. Guides asynchronous processing, where waiting tasks do not block execution, plus language model connections, external tools, workflow coordination, error handling, and tests.
Apply related edits across several files using a unified diff, a file describing code changes. Check the patch with Git before applying it, and rebuild rejected patches using the current file contents.
Turn analysis results into charts suited to the question, such as trends, comparisons, or distributions. Save a chart image and the script that generated it, with a short explanation of the findings and relevant caveats.
Locate function and class definitions, usages, and relevant files with ripgrep, using text patterns and file filters. Search Git history to see when text appeared or disappeared and investigate who changed specific lines.
Calculate the count, sum, minimum, maximum, and average of supplied numbers or datasets, then present the results in a clear, structured format. Analysis results can also be saved to a file when requested.
Use a starter template to learn how an agent skill is organized and outline its intended use, required inputs, steps, and output format. Replace the placeholder body with your own working instructions before relying on it.