arize-prompts

Pass

Audited by Gen Agent Trust Hub on Sep 11, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the ax command-line interface to perform prompt management, version control, and space configuration tasks.
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions for installing the official arize-ax-cli package from standard registries using pip, pipx, or uv.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses a vulnerability surface where it ingests untrusted data from LLM traces.
  • Ingestion points: 'SKILL.md' Workflow B uses 'ax spans export' to retrieve external message content.
  • Boundary markers: The skill includes explicit warnings to treat exported span text as untrusted.
  • Capability inventory: The agent can execute CLI commands and write to local files (e.g., 'messages.json').
  • Sanitization: The skill requires the agent to confirm reconstructed messages with the user before proceeding.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 11, 2026, 12:20 AM
Security Audit — agent-trust-hub — arize-prompts