Skills
skills/k-dense-ai/claude-scientific-skills/reactome-database/Gen Agent Trust Hub

reactome-database

Pass

Audited by Gen Agent Trust Hub on Mar 4, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill utilizes the reactome2py Python package (v3.0.0) from PyPI for programmatic access to the database. \n- [COMMAND_EXECUTION]: Provides a helper script scripts/reactome_query.py that utilizes the requests library to communicate with Reactome REST API endpoints. \n- [DATA_EXFILTRATION]: Transmits user-provided gene or protein lists to the Reactome Analysis Service (https://reactome.org) for enrichment calculations. These network operations target a well-known and established scientific domain. \n- [PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection when processing external data files: \n
  • Ingestion points: The scripts/reactome_query.py script reads gene identifiers from local text files via the analyze command. \n
  • Boundary markers: No specific delimiters or instructions are used to distinguish user data from potential instructions within the input files. \n
  • Capability inventory: The skill uses the requests library in scripts/reactome_query.py for network operations. \n
  • Sanitization: Input data is read and used in API requests without strict schema validation or escaping.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 4, 2026, 07:49 AM