chembl-database

Pass

Audited by Gen Agent Trust Hub on Oct 8, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill recommends the installation of the chembl_webresource_client and pandas packages. These are standard, well-known libraries for bioinformatic analysis and accessing the official EBI ChEMBL database.
  • [INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection as it retrieves and processes data from the external ChEMBL API within the agent's context.
  • Ingestion points: Data is ingested via new_client filter methods and search functions across SKILL.md and scripts/example_queries.py (e.g., search_molecules_by_name, get_bioactivity_data).
  • Boundary markers: No specific delimiters or safety instructions are used to wrap the data retrieved from the external API.
  • Capability inventory: The skill possesses network access capabilities to query multiple ChEMBL database endpoints (molecules, targets, activities, etc.).
  • Sanitization: No specific sanitization or filtering of the external API response is implemented before the content is presented to the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 8, 2026, 09:44 PM