axiom-ios-vision
Pass
Audited by Gen Agent Trust Hub on Feb 23, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [No Code] (SAFE): The analyzed file is a markdown-based routing document. It contains instructions, decision logic, and example invocations but does not include any executable scripts, shell commands, or binary assets.
- [Prompt Injection] (SAFE): No instructions attempting to override the agent's core behavior, bypass safety filters, or extract system prompts were detected.
- [Data Exposure & Exfiltration] (SAFE): The skill does not access sensitive local file paths (e.g., SSH keys, credentials) and contains no network-capable code to transmit data externally.
- [Obfuscation] (SAFE): No hidden characters, Base64 encoding, or homoglyph-based evasion techniques were found in the text.
- [External Downloads] (SAFE): There are no links to external scripts, package managers, or remote repositories.
Audit Metadata