The Agent Skills Directory

[CREDENTIALS_UNSAFE]: The skill follows security best practices by instructing the use of environment variables (os.environ['POSTBRIDGE_API_KEY']) for API authentication rather than hardcoding secrets within the code or prompts.
[COMMAND_EXECUTION]: The skill includes a Python code snippet that uses the requests library to interact with external APIs. The endpoint used in the example, https://your-crosspost-service.example/api/posts, utilizes the .example top-level domain, which is reserved for documentation and testing, posing no risk of unauthorized data transmission to live external servers.
[PROMPT_INJECTION]: The skill acts as a distribution engine for user-provided content. While this creates a potential surface for indirect prompt injection (where malicious instructions embedded in source content could attempt to influence the agent's behavior), the skill includes specific 'Core Rules' and a 'Quality Gate' checklist designed to ensure content is adapted and validated before being sent to social media APIs.

crosspost