The Agent Skills Directory

[SAFE]: Analysis of the skill instructions and metadata found no evidence of prompt injection, obfuscation, or malicious patterns.
[SAFE]: The included script scripts/topic_cluster_mapper.py is written in clean Python and relies exclusively on built-in modules (argparse, json, re, sys, collections). It avoids risky functions like eval() or exec() and does not perform any network or unauthorized file system operations.
[SAFE]: The skill does not attempt to download external code, escalate privileges, or establish persistence on the host system.
[SAFE]: Although the skill is designed to process external inputs like customer transcripts and keyword exports, it possesses no outbound capabilities (such as network access) that would allow for data exfiltration or exploitation through indirect prompt injection.

content-strategy