The Agent Skills Directory

[SAFE]: No malicious patterns, obfuscation, or unauthorized data access attempts were detected. The skill's operations are strictly confined to reading and writing specific data columns within the user's workspace as described.
[SAFE]: The skill demonstrates a high level of defensive engineering by requiring a human-review gate for sample data and implementing six specific code-based guards (e.g., checking for acronyms, non-Latin scripts, and hallucinatory output) to prevent incorrect data modification.
[SAFE]: The instructions explicitly forbid network exfiltration, automated messaging, or CRM writes outside the defined columns, significantly limiting the potential impact of indirect prompt injection or other logic manipulation.

clean-first-names-for-greetings