nearby-species

Pass

Audited by Gen Agent Trust Hub on Jun 19, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: No security issues, malicious code, or suspicious behavior patterns were identified. The skill instructions are limited to data formatting and display logic.
  • [PROMPT_INJECTION]: The skill instructions define an interface for Indirect Prompt Injection through the processing of third-party species data.
  • Ingestion points: External biodiversity data retrieved from the iNaturalist database via the nearby_species tool.
  • Boundary markers: None provided in the instruction set.
  • Capability inventory: The skill does not utilize tools for file system access, command execution, or network exfiltration.
  • Sanitization: No input validation or sanitization of tool outputs is performed.
  • Risk Level: The risk is negligible as the skill lacks any capabilities that would allow an attacker to exploit the processed data.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 19, 2026, 04:41 AM
Security Audit — agent-trust-hub — nearby-species