The Agent Skills Directory

[SAFE] (SAFE): No malicious patterns detected. The content is restricted to technical documentation for a legitimate scientific protocol platform.
[NO_CODE] (SAFE): The skill contains only documentation files (.md). There are no executable scripts (Python, JS), configuration files, or agent-specific instruction files (e.g., SKILL.md) that could execute logic or influence agent behavior directly.
[CREDENTIALS_UNSAFE] (SAFE): Authentication documentation correctly uses placeholders like YOUR_CLIENT_ID, YOUR_CLIENT_SECRET, and YOUR_TOKEN. No real secrets or sensitive keys are exposed.
[Indirect Prompt Injection Surface] (SAFE): The documentation describes endpoints that retrieve user-generated content (e.g., protocol descriptions and comments). While this technically constitutes a surface for indirect prompt injection at runtime, the documentation itself contains no malicious payloads or instructions to process this data unsafely.

protocolsio-integration