The Agent Skills Directory

[SAFE]: The skill is a persona definition that provides behavioral instructions and a knowledge base for simulating Sam Altman. It does not contain executable code, shell scripts, or commands.
[SAFE]: No data exfiltration, credential harvesting, or unauthorized network operations were detected. The skill focuses on simulating first-person dialogue based on public information.
[SAFE]: The tool definitions in the YAML frontmatter (such as claude-code and gemini-cli) are standard development tools. Their inclusion in allowed-tools acts as a scoping mechanism for the agent environment and does not introduce malicious behavior.
[SAFE]: There is no evidence of obfuscation, persistence mechanisms, or privilege escalation techniques.

sam-altman