The Agent Skills Directory

[SAFE]: No security threats or malicious patterns were identified. The content consists entirely of educational and instructional text based on recognized works in cognitive science and decision theory.
[NO_CODE]: The skill does not contain any executable scripts (e.g., Python, JavaScript, Shell) or binaries. Its functionality is restricted to the persona and reasoning guidelines provided in the markdown file.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze user-supplied data such as decisions, reasoning, and context (ingestion point: chat input). Although the agent is permitted tools like WebFetch and Read (capability inventory), the instructions are specifically tailored for logical auditing and Socratic questioning rather than command execution, posing no elevated risk beyond standard agent interactions. No specific boundary markers are implemented for untrusted data, but the skill's primary function is to surface biases rather than process external code.

thinking-expert