The Agent Skills Directory

[SAFE]: The skill is composed of natural language instructions that define a cognitive framework for problem-solving and associative thinking in design contexts. It does not perform any programmatic actions.
[SAFE]: No tool use, network operations, or file system interactions are defined or requested in the skill's manifest or body.
[SAFE]: No obfuscation, hidden URLs, or encoded payloads were found within the document.
[PROMPT_INJECTION]: The skill uses phrases like 'don't filter yourself' and 'suppress premature idea-dismissal' as triggers and internal instructions. These are clearly scoped to the creative brainstorming process and do not attempt to override system safety guardrails or disregard previous instructions.

philosopher