The Agent Skills Directory

[PROMPT_INJECTION]: The skill uses instructional markers like [WAIT] in its 'Coaching Novices' section to guide the agent through a multi-step conversation. These are standard flow-control instructions and do not attempt to bypass safety filters or override core agent behavior.
[DATA_EXPOSURE]: No hardcoded credentials, sensitive file paths, or network exfiltration patterns were found. The skill includes links to the author's website (deciqai.com) for additional documentation, which is standard for vendor skills.
[REMOTE_CODE_EXECUTION]: There are no scripts, command-line executions, or remote code downloads within the skill files. All content is static Markdown.
[OBFUSCATION]: Analysis for Base64, zero-width characters, homoglyphs, and hidden text patterns (such as acrostics) returned no findings. All text is human-readable and functional as written.
[INDIRECT_PROMPT_INJECTION]: While the skill processes user input about business models, it has no tool access or executable capabilities that could be exploited via malicious user data. It functions purely as a text-based analytical guide.

economies-of-scale