frida-codeshare-search

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.95). The required workflow fetches public Frida CodeShare HTML at runtime (e.g., curl ... https://codeshare.frida.re/search/ and curl ... https://codeshare.frida.re/@.../), then the agent parses/extracts readable fields including projectSource from that outsider-authored page into the LLM context for review/execution.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 1.00). The skill explicitly shows runtime fetching and execution of remote Frida CodeShare scripts (e.g., curl "https://codeshare.frida.re/@Eltion/instagram-ssl-pinning-bypass/" and the direct execution command frida --codeshare Eltion/instagram-ssl-pinning-bypass) which would download and run third‑party code at runtime.