The Agent Skills Directory

[SAFE]: The skill acts as an architectural playbook and reference guide for the Agently framework. It does not execute code or perform network operations itself, but rather provides instructions on how to structure projects.
[DATA_EXPOSURE]: The instructions explicitly promote secure credential management. It advises users to use environment variable placeholders (e.g., ${ENV.xxx}) and .env files instead of hardcoding secrets or API keys in configuration files or codebases.
[PROMPT_INJECTION]: The skill provides guidance on prompt composition and management but does not include any instructions that attempt to bypass safety filters, override system behavior, or extract sensitive instructions.
[COMMAND_EXECUTION]: The skill describes the use of tools and extensions (e.g., FastAPIHelper, MCP, search, browse) but does not implement or invoke them directly in a way that allows for arbitrary command injection.

agently-playbook