2.5k
GitHub Stars
5
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill openclaw/skills --skill call-center- _meta.json279 B
- escalation.md3.0 KB
- memory-template.md2.3 KB
- scripts.md3.3 KB
- SKILL.md3.3 KB
Overview
This skill provides a full call center agent workflow for handling voice interactions, including scripts, issue resolution steps, escalation protocols, and interaction logging. It’s designed to standardize greetings, verification, troubleshooting, documentation, and post-call work to improve consistency and measurable outcomes. The skill stores agent memory and artifacts locally for quick reference and analytics.
How this skill works
The skill inspects incoming call context and selects the appropriate script (support, sales, collections) to guide the agent through a structured flow: greet, verify, listen, act, and close. It enforces core rules such as identity verification, active listening, escalation triggers, and logging. Interaction data and metrics are kept in a local memory structure to support reporting, pattern detection, and continuous improvement.
When to use it
- Inbound customer support calls needing troubleshooting or account help
- Outbound campaigns for sales or collections following a script
- Situations that may require supervisor escalation or legal/compliance attention
- Training new agents on standard greeting, verification, and documentation practices
- Monitoring and improving call metrics like FCR, CSAT, and after-call work
Best practices
- Always verify caller identity before discussing account information to avoid compliance risks
- Listen fully and paraphrase the issue before proposing solutions to reduce repeat calls
- Document the reason, actions, outcomes, and any promises immediately after the call (aim <2 minutes)
- Escalate on explicit request, authority limits, legal mentions, or serious safety concerns
- Avoid over-promising timeframes; set realistic commitments and record deadlines
Example use cases
- A Tier 1 agent follows the support script to diagnose and resolve a billing dispute, logging steps and outcome
- A sales team runs an outbound campaign using the sales script to qualify leads and schedule demos
- Collections agents use the collections flow to verify balances, offer payment options, and document agreements
- Supervisors review escalations.md to triage issues that exceeded agent authority or flagged compliance risk
- Quality team analyzes metrics.md to target coaching for low FCR and long handle times
FAQ
Memory and artifacts are stored in the ~/call-center/ directory, including memory.md, scripts/, escalations.md, and metrics.md.
When should I escalate a call?
Escalate immediately if the caller requests a supervisor, the issue is outside authority, legal/compliance is mentioned, safety is at risk, or after three failed attempts to resolve.