The Agent Skills Directory

[SAFE]: The skill acts exclusively as a set of instructions for an AI agent to perform logical reasoning. It does not include any functional code or system commands.
[NO_CODE]: No scripts, binaries, or external dependencies are defined or utilized within the skill.
[PROMPT_INJECTION]: The skill is designed to process external inputs (generated content and source material). This creates a surface for indirect prompt injection where the input data could contain instructions to manipulate the agent's scoring. However, since the skill has no access to the file system, network, or shell, the security risk is negligible, and the impact is limited to the accuracy of the fact-check result.

factuality-check