The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill interacts exclusively with the official protocols.io API endpoints at https://protocols.io/api/v3. This is a well-known scientific service, and all network activity is strictly limited to the platform's core functionalities as described in the documentation.
[COMMAND_EXECUTION]: The skill provides Python implementation examples using the standard requests library and Bash examples using curl. These snippets are intended for API communication and do not contain patterns for arbitrary command execution or shell injection.
[CREDENTIALS_UNSAFE]: Security best practices are followed throughout the documentation. Implementation examples use generic placeholders such as YOUR_ACCESS_TOKEN and YOUR_CLIENT_ID, and the instructions explicitly warn against hardcoding secrets or exposing them in version control.
[PROMPT_INJECTION]: While the skill retrieves user-generated content from protocols.io (such as comments and protocol steps), the data ingestion occurs through a controlled API environment. The risk of indirect prompt injection is minimized by the specialized scientific nature of the data and the absence of instructions that would cause the agent to execute instructions found within the data.

protocolsio-integration