elife

Pass

Audited by Gen Agent Trust Hub on Sep 24, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill requires the agent to ingest data from external sources, creating a potential surface for indirect prompt injection.
  • Ingestion points: Instructions in SKILL.md direct the agent to read local resource files (../../resources/source-basis.md, ../../resources/official-source-map.md) and search elifesciences.org for live author instructions.
  • Boundary markers: Absent. There are no instructions to use delimiters or to treat the ingested external content as non-authoritative data.
  • Capability inventory: The skill does not include capabilities for arbitrary command execution, network exfiltration, or sensitive file writing, which limits the risk profile.
  • Sanitization: No validation or sanitization routines are specified for the external content before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 24, 2026, 11:33 AM
Security Audit — agent-trust-hub — elife