discover-x402-services
Fail
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: HIGHEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS] (HIGH): The skill uses
npxto download and run the@zauthx402/clipackage from the public npm registry at runtime. This package is not from a trusted source, creating a significant supply chain vulnerability. - [REMOTE_CODE_EXECUTION] (HIGH): The use of
npx @zauthx402/cli@latestfacilitates the execution of unverified remote code within the agent's execution environment. - [COMMAND_EXECUTION] (MEDIUM): The skill interpolates user-provided search queries directly into a bash command string (
npx ... search <query>), which presents a potential command injection risk. - [PROMPT_INJECTION] (HIGH): The skill possesses a major indirect prompt injection surface. 1. Ingestion points: It fetches and displays
titleanddescriptionfields from the external zauth provider directory. 2. Boundary markers: No delimiters or instructions are used to isolate this untrusted external data from the agent's primary instructions. 3. Capability inventory: The skill has access to theBashshell and is intended to trigger subsequent actions like payments via other skills. 4. Sanitization: There is no evidence of sanitization or filtering applied to the external API metadata before it is processed by the AI agent.
Recommendations
- AI detected serious security threats
Audit Metadata