The Agent Skills Directory

[SAFE]: The skill outlines integration patterns for the Ollama API, specifically targeting local communication via localhost:11434, which is the standard default for this service.
[SAFE]: Documentation explicitly highlights security best practices, such as warning against exposing the raw Ollama port to the internet and instructing the user to avoid logging prompts containing secrets or personally identifiable information (PII).
[SAFE]: Implementation examples use standard tools like curl and the TypeScript fetch API without introducing any malicious command execution, unauthorized file system access, or persistence mechanisms.
[SAFE]: No obfuscation, remote code execution from untrusted sources, or prompt injection patterns were detected within the skill's instructions or metadata.

ollama-api