The Agent Skills Directory

[SAFE]: The skill provides structured instructions and best practices for creating social media graphics tailored to specific platforms like Instagram, YouTube, and LinkedIn.
[SAFE]: No suspicious network operations, sensitive file access, or persistence mechanisms were identified in the instructions.
[SAFE]: The use of the external tool Glif with a specific workflow ID is consistent with the skill's stated purpose of image generation and does not involve arbitrary remote code execution.
[PROMPT_INJECTION]: The skill utilizes user-provided input by interpolating a 'subject' into prompt templates (e.g., [Subject] for Instagram feed). This creates a surface for indirect prompt injection where an attacker could provide a malicious subject to influence the generated image. However, since the output is a static graphic and not executable code or text-based logic, the security risk is negligible.
Ingestion points: SKILL.md (multiple prompt templates using the [Subject] placeholder)
Boundary markers: Absent
Capability inventory: Image generation via external Glif API
Sanitization: Absent

ai-social-graphics