The Agent Skills Directory

[SAFE]: No security issues were identified during the analysis. The skill is designed as an educational assistant and its behavior is strictly confined to providing information based on its internal reference files.
[NO_CODE]: The skill consists exclusively of Markdown files providing instructional content. It does not include any executable scripts, binaries, or automated configuration changes.
[INDIRECT_PROMPT_INJECTION]: The skill allows the agent to ingest and analyze student-provided descriptions of artwork and annotations for feedback.
Ingestion points: Identified in SKILL.md (Core Workflow section) where the agent asks to see or have work described.
Boundary markers: Not explicitly defined in the prompts.
Capability inventory: The skill lacks any capability to execute tools, make network requests, write files, or spawn subprocesses.
Sanitization: No explicit sanitization is described, however, the total lack of external capabilities renders the attack surface non-exploitable.
[METADATA_POISONING]: All metadata fields, including the description and exam-specific tags, accurately reflect the skill's stated purpose and contain no deceptive or malicious instructions.

gcse-art-tutor