reactome-database

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes biological data and reports from the Reactome API, which could potentially contain malicious instructions meant to influence the agent's behavior.\n
  • Ingestion points: Data retrieved via the reactome_analysis.py tool from the Reactome Analysis and Content Services (reactome.org), including JSON, CSV, and PDF responses.\n
  • Boundary markers: Absent. The skill instructions do not define specific delimiters for external data or specify that the agent should ignore instructions found within the API output.\n
  • Capability inventory: The associated Python script has capabilities for reading and writing local files and performing network requests restricted to the Reactome domain.\n
  • Sanitization: The script performs statistical filtering (e.g., FDR and p-value) but does not sanitize text content for potentially malicious prompt injection patterns.\n- [EXTERNAL_DOWNLOADS]: The skill script identifies a dependency on the polite-http library. This package is downloaded from the standard Python registry during environment setup and is used for rate-limited interaction with the Reactome API.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 07:46 AM
Security Audit — agent-trust-hub — reactome-database