papers

Pass

Audited by Gen Agent Trust Hub on May 11, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns were detected. The skill utilizes established academic APIs (arXiv, Semantic Scholar, PubMed) and focuses on document analysis for research purposes, which is consistent with its stated intent.
  • [PROMPT_INJECTION]: The skill processes untrusted external PDF content, which represents a surface for indirect prompt injection. This risk is inherent to the skill's primary purpose of parsing third-party research documents and is generally managed by the underlying model's safety guardrails.
  • Ingestion points: External PDFs are fetched from sources including arXiv, Semantic Scholar, Unpaywall, CORE, and PubMed as defined in SKILL.md.
  • Boundary markers: Absent; the instructions do not define explicit delimiters or instructions for the agent to disregard commands embedded within the fetched document text.
  • Capability inventory: The skill uses the leyline:document-conversion protocol and extraction targets to convert PDFs to markdown and summarize findings.
  • Sanitization: Absent; the skill does not specify any validation or sanitization steps for the data extracted from the external documents.
Audit Metadata
Risk Level
SAFE
Analyzed
May 11, 2026, 02:36 AM