web-auth-bypass

Warn

Audited by Gen Agent Trust Hub on Sep 19, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the agent to download and install the jwt_tool utility from an external GitHub repository (https://github.com/ticarpi/jwt_tool.git) which is not associated with a trusted organization or well-known service. It also lists several non-standard Python and Node.js packages as requirements.
  • [INDIRECT_PROMPT_INJECTION]: The skill operates by fetching and processing untrusted data from external web applications using curl, requests, and WebFetch. This creates a vulnerability surface for indirect prompt injection attacks where the processed data could contain instructions targeting the agent.
  • Ingestion points: Responses from web applications targeted for testing as seen in SKILL.md, payloads.md, and various guides.
  • Boundary markers: Not present; the skill lacks delimiters or instructions for the agent to ignore commands within the fetched content.
  • Capability inventory: The skill uses Bash for network operations and Write for managing tokens and scan results locally.
  • Sanitization: No evidence of filtering or escaping external data before it is integrated into the agent's workflow.
  • [COMMAND_EXECUTION]: The skill provides numerous Bash and Python scripts designed for the agent to perform network-based attacks. These scripts involve automated login attempts and parallel requests to external network services.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Sep 19, 2026, 04:36 PM