agentic-data-scientist

Pass

Audited by Gen Agent Trust Hub on May 18, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The installation instructions fetch the agentic-data-scientist package via uv and the @anthropic-ai/claude-code package from the official npm registry. These originate from the vendor or trusted organizations.
  • [COMMAND_EXECUTION]: The skill uses a command-line interface to perform data operations, file management, and agent orchestration, which is the intended purpose of a data science framework.
  • [PROMPT_INJECTION]: The skill represents an indirect prompt injection surface as it processes external datasets and natural language queries.
  • Ingestion points: User queries and input files (CSV, JSON, etc.) ingested via the --files parameter.
  • Boundary markers: None identified in the skill documentation.
  • Capability inventory: The coding agent can execute arbitrary scientific code through the claude-code CLI and local libraries.
  • Sanitization: There are no documented sanitization steps for the data processed by the agents.
Audit Metadata
Risk Level
SAFE
Analyzed
May 18, 2026, 12:56 AM
Security Audit — agent-trust-hub — agentic-data-scientist