arize-ai-provider-integration

Pass

Audited by Gen Agent Trust Hub on Apr 2, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the ax (Arize CLI) tool for all operations, including listing, creating, and updating integrations. It also provides commands to modify user shell configuration files (~/.zshrc, ~/.bashrc) or Windows environment variables to persist settings like ARIZE_SPACE_ID.\n- [EXTERNAL_DOWNLOADS]: The references/ax-setup.md file provides instructions to install or upgrade the arize-ax-cli package from public registries using pip, uv, or pipx.\n- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it processes data (such as integration names and metadata) retrieved from the Arize platform. An attacker with control over integration names in the Arize UI could potentially influence the agent's behavior when it lists or gets integration details.\n
  • Ingestion points: Output from ax ai-integrations list and ax ai-integrations get in SKILL.md.\n
  • Boundary markers: None; integration data is used directly in the conversation context.\n
  • Capability inventory: Command execution (ax), system configuration modification (shell profiles), and credential management.\n
  • Sanitization: No validation or sanitization of integration metadata is performed before the agent processes it.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 2, 2026, 01:04 AM