axiom-ios-ai
Pass
Audited by Gen Agent Trust Hub on Feb 23, 2026
Risk Level: SAFE
Full Analysis
- [Prompt Injection] (SAFE): No instruction override, role-play, or safety bypass patterns were detected. The routing instructions are purely functional.
- [Data Exposure & Exfiltration] (SAFE): No sensitive file paths (e.g., ~/.ssh, ~/.aws), hardcoded secrets, or network exfiltration patterns were found.
- [Remote Code Execution] (SAFE): The skill does not perform any package installations or remote script executions. It is a configuration-only skill.
- [Indirect Prompt Injection] (SAFE):
- Ingestion points: User input is analyzed by the router in SKILL.md to map queries to sub-skills.
- Boundary markers: Not present, as the file defines logic rather than an interactive prompt shell.
- Capability inventory: No dangerous capabilities (subprocess calls, file-write, or network operations) exist in this skill.
- Sanitization: None; however, the skill has no exploitable sinks for malicious data.
Audit Metadata