search-skills
Pass
Audited by Gen Agent Trust Hub on Nov 9, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill set serves as a legitimate documentation and reference library for academic research workflows. No malicious patterns such as prompt injection, unauthorized data exfiltration, or hidden persistence mechanisms were detected.
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the retrieval of external data from scholarly databases. While these sources are reputable, the processed content (abstracts, full texts) represents a potential attack surface for indirect prompt injection if an attacker were to publish a paper containing malicious instructions. This risk is inherent to any literature search capability.
- Ingestion points: API endpoints for arXiv, PubMed, OpenAlex, and other academic services as documented in the respective
SKILL.mdfiles. - Boundary markers: No explicit delimiters or instructions to ignore embedded commands are provided for handling the retrieved text.
- Capability inventory: The guides describe the use of network tools (
curl,requests) to fetch data and processing tools to analyze academic text. - Sanitization: No specific sanitization or filtering logic is mentioned for the ingested academic content.
Audit Metadata