The Agent Skills Directory

[SAFE]: The skill is a collection of Markdown files defining a 'Caveman Mode' persona for the agent. It does not perform any operations or request any permissions beyond stylistic formatting.
[NO_CODE]: No scripts, binaries, or configuration files capable of execution are included in the skill package.
[PROMPT_INJECTION]: Instructions are focused on response formatting and token efficiency. No instructions were found that attempt to bypass AI safety guidelines, disregard prior instructions for malicious purposes, or manipulate core agent behavior beyond response style.
[DATA_EXFILTRATION]: There are no indicators of network communication, use of external APIs, or access to sensitive local data stores.
[OBFUSCATION]: Content was scanned for Base64, zero-width characters, homoglyphs, and other encoding methods; no obfuscation techniques were detected.

caveman