The Agent Skills Directory

[SAFE]: The skill handles sensitive information, such as Arize API keys, according to security best practices by directing the user to use environment variables rather than providing secrets directly to the agent.\n- [PROMPT_INJECTION]: The skill correctly identifies a potential indirect prompt injection surface in exported trace data (e.g., span attributes) and provides explicit instructions for the agent to treat this data as untrusted text, effectively mitigating the risk. Evidence: File SKILL.md contains a Security Guardrail section warning about untrusted user-generated content in span attributes.\n- [COMMAND_EXECUTION]: The skill executes the ax CLI tool to fetch trace data from official Arize domains. All command operations are within the scope of the skill's intended purpose for telemetry analysis.\n- [EXTERNAL_DOWNLOADS]: The skill references the arize-ax-cli package and provides troubleshooting instructions for its installation from official registries via standard package managers.

arize-trace