The Agent Skills Directory

[SAFE]: The skill is entirely instructional and provides strategic frameworks for YouTube content creation. Technical analysis of all 13 files confirms the absence of malicious intent or dangerous capabilities.
[COMMAND_EXECUTION]: The skill utilizes two local Python scripts, scripts/ctr-baseline.py and scripts/title-check.py. These scripts are benign, using only standard Python libraries (argparse, json, re, sys) to perform mathematical calculations and regex-based heuristic analysis. No unsafe functions such as eval(), exec(), or os.system() are used.
[PROMPT_INJECTION]: No evidence of prompt injection, safety bypasses, or instructions to ignore system guidelines was detected. The instructions are focused on guiding the AI agent to follow a specific planning methodology.
[DATA_EXFILTRATION]: No network operations (curl, wget, fetch) or sensitive file access patterns were found. The skill operates within its own context and processes user-provided inputs for video planning.
[OBFUSCATION]: A scan for Base64, zero-width characters, Unicode tags, and homoglyphs returned negative results. All content is in plain text and human-readable.
[DYNAMIC_CONTEXT_INJECTION]: The skill does not use the !command syntax or any other mechanism to execute shell commands at load time. Commands mentioned in the documentation are intended for the agent to run manually as part of the analysis workflow.
[METADATA_POISONING]: The skill metadata (author: Pedro Nauck) and descriptions are consistent with the instructional content and the provided author context.

yt-master