The Agent Skills Directory

[SAFE]: The skill is a reference-only resource containing documentation, methodology principles, and document templates. It does not include any scripts, executable code, or third-party dependencies. All external references are to well-known technology blogs and documentation sites for the purpose of attribution and methodology explanation.
[INDIRECT_PROMPT_INJECTION]: The skill defines patterns for agents to process untrusted user input, such as product ideas and feature requests. However, the skill lacks any tools or scripts with system-level capabilities (such as file system access or network operations), eliminating the risk of a dangerous capability chain. The methodology specifically encourages structured narrowing of problems, which provides a natural logical filter for user-provided data.

product-shaping