explore-dataset
Pass
Audited by Gen Agent Trust Hub on Apr 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is authored by axiomhq and utilizes the official axiom CLI, representing standard vendor functionality for dataset management and observability. All external interactions are conducted through this authenticated tool.
- [COMMAND_EXECUTION]: The skill uses shell-based commands via the axiom CLI (e.g., axiom query, axiom dataset list) to perform data discovery. These operations are within the scope of the defined allowed-tools and the skill's intended purpose.
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it processes external log data from Axiom datasets. 1. Ingestion points: Data returned from axiom query commands in SKILL.md. 2. Boundary markers: Absent. 3. Capability inventory: Shell execution via the axiom CLI. 4. Sanitization: None; content is processed directly by the agent for analysis. This risk is inherent to log analysis tools and is documented as a baseline security consideration.
Audit Metadata