skills/crtvrffnrt/skills/pentest-xss/Gen Agent Trust Hub

pentest-xss

Pass

Audited by Gen Agent Trust Hub on Feb 24, 2026

Risk Level: SAFEDATA_EXFILTRATIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTIONNO_CODE
Full Analysis
  • [DATA_EXFILTRATION]: The skill provides example payloads for exfiltrating sensitive information, such as browser cookies, to an external server in examples/match-and-replace-examples.md.
  • [COMMAND_EXECUTION]: Includes XSS payloads that use eval() for dynamic execution on a target's browser to bypass security controls in examples/match-and-replace-examples.md.
  • [EXTERNAL_DOWNLOADS]: Recommends using the well-known interact.sh service for Out-of-Band interaction testing in resources/blind-xss-guide.md.
  • [PROMPT_INJECTION]: Potential for indirect prompt injection as the skill analyzes untrusted data from web applications. Ingestion points include user-controllable input vectors and web headers (SKILL.md). No boundary markers or sanitization procedures for handling untrusted data are specified in the instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 24, 2026, 12:36 PM