The Agent Skills Directory

[SAFE]: The instructions provided in SKILL.md are focused on content marketing and strategy without any signs of malicious intent, prompt injection, or safety bypass attempts.\n- [SAFE]: The utility script scripts/topic_cluster_mapper.py performs local keyword clustering using standard Python libraries (argparse, json, re, sys, collections). It does not perform network operations, use unsafe dynamic execution (e.g., eval or exec), or require third-party dependencies.\n- [SAFE]: No obfuscation, persistence mechanisms, privilege escalation, or unsafe credential management was detected in the skill configuration or code.\n- [SAFE]: The skill processes untrusted data sources (keyword exports, transcripts, forum posts) to generate content ideas, which represents a potential surface for indirect prompt injection.\n
Ingestion points: SKILL.md (Ideation sources: Keyword Data, Call Transcripts, Survey Responses, Forum Research).\n
Boundary markers: Absent. The instructions do not specify the use of delimiters or 'ignore' instructions for the external content.\n
Capability inventory: Web search and script execution (scripts/topic_cluster_mapper.py).\n
Sanitization: Absent. Content is processed directly for theme extraction and analysis.

content-strategy