setup-cell2location-remote-tool

Pass

Audited by Gen Agent Trust Hub on Oct 11, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches configuration and dependencies from 'connect.aiscientist.tools' and the 'mims-harvard/ToolUniverse' GitHub repository. It also communicates with a backend service hosted on 'onrender.com'.
  • [COMMAND_EXECUTION]: Uses shell commands including 'uv', 'pip', and 'tu' to create Python virtual environments, install packages, and manage the tool's execution and sharing.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external data which creates a vulnerability surface.
  • Ingestion points: The 'run_cell2location_deconvolution' tool in 'SKILL.md' accepts 'sc_path' and 'sp_path' arguments.
  • Boundary markers: No specific delimiters or safety instructions are defined for the content of these data files.
  • Capability inventory: The skill is capable of network communication and file system access for data processing.
  • Sanitization: There is no evidence of content validation or sanitization for the ingested scientific data files.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 11, 2026, 05:27 PM