Customer Churn Prediction Analyst

Key points to consider before installing or running this skill: - Data exfiltration risk: The skill requires OPENAI_API_KEY, which means customer data could be sent to OpenAI for analysis or text generation. Do not provide full PII/payment data unless you are comfortable with that external processing and have legal/contractual approval. - Least privilege: If you proceed, create and use restricted API keys where possible (Stripe restricted key with read-only scopes, Shopify token limited to necessary endpoints) and rotate keys after use. - Test with anonymized samples: Run the skill on scrubbed or synthetic customer data first to observe what is sent to external endpoints and what outputs are produced. - Confirm missing credentials flow: The SKILL.md mentions Slack and other SaaS integrations but does not request Slack/CRM credentials. Ask the publisher how those integrations are implemented and whether the skill will prompt you for additional tokens or expects file uploads. - Audit and logging: Ensure you can audit API calls (Stripe/Shopify logs) to detect unexpected access. Avoid putting long-lived high-privilege keys into the environment. - Source verification: The skill lists a GitHub homepage but 'Source: unknown' in the registry. Review the linked repo and author, and prefer skills with transparent maintainers and code available. If you cannot confirm how data is handled (redaction, retention, third-party transmission) or cannot limit credentials, treat this skill as high-risk and do not provide production customer data.