The Agent Skills Directory

[SAFE]: The skill performs legitimate configuration management by reading and writing a local file (p3-models.md) to store model preferences. It employs security best practices by validating all user-selected models against the system's orchestrator_capabilities result before writing them to the configuration, ensuring that only authorized and existing models can be configured. No instances of prompt injection, data exfiltration, obfuscation, or unauthorized command execution were found.

setup-p3