The Agent Skills Directory

[SAFE]: The skill consists of instructional markdown and templates for customer success practices. No malicious code, suspicious commands, or network exfiltration patterns were found within the provided files. The content is purely instructional and logic-based for the agent's operational workflow.- [SAFE]: The skill includes comprehensive 'Privacy and Human Judgment Boundaries' which mandate human-in-the-loop reviews for all actions and provide guidance on data minimization and consent. These guidelines act as effective guardrails against the automated misuse of customer data and mitigate risks associated with indirect prompt injection in customer feedback.- [SAFE]: Precondition assessments and applicability decisions are used to ensure the skill only operates in appropriate contexts. All routing to other skills (such as product-adoption or go-to-market) follows established catalog patterns and does not use unsafe dynamic loading or execution mechanisms.

conditional-customer-success