The Agent Skills Directory

[COMMAND_EXECUTION]: The skill invokes local utility scripts (calculate_scores.py and generate_schematic.py) to process scoring data and generate visual representations. These are legitimate analytical operations intended for academic work evaluation.
[INDIRECT_PROMPT_INJECTION]: The skill contains an attack surface for indirect prompt injection as it ingests user-provided research papers and proposals (SKILL.md workflow). While it lacks explicit boundary markers and input sanitization, the risk is considered low due to the structured, dimension-based evaluation methodology. The skill's capabilities are limited to local file reporting through its provided scripts.
[SAFE]: No evidence of obfuscation, remote code execution, or sensitive data exfiltration was detected. The skill relies on standard Python libraries and references peer-reviewed academic frameworks.

scholar-evaluation