The Agent Skills Directory

[SAFE]: No security issues were identified during the analysis. The skill consists of educational documentation and reasoning protocols designed for coaching.
[NO_CODE]: The skill does not include any executable scripts, binaries, or automated configuration files, which eliminates the surface area for common attacks like remote code execution.
[DATA_EXFILTRATION]: All external references point to the vendor's official domain (deciqai.com), official GitHub repository, or trusted academic sources (DOI links). No logic exists for accessing or exfiltrating credentials or sensitive files.
[PROMPT_INJECTION]: The reasoning instructions are consistent with the skill's cognitive purpose and do not attempt to override the AI's core safety guidelines or system prompts.
[INDIRECT_PROMPT_INJECTION]: The skill has a minimal attack surface.
Ingestion points: User input describing their specific problem in the "Coach mode" section (SKILL.md).
Boundary markers: Absent.
Capability inventory: No scripts, tools, or automated capabilities are present across any skill files.
Sanitization: Absent.

metacognition