web-vuln-mfa-bypass
Pass
Audited by Gen Agent Trust Hub on Aug 19, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes shell commands using
curlandparallelto perform network probes, stress tests, and race condition analysis against user-defined web endpoints. This behavior is consistent with the skill's stated purpose as a security testing tool. - [INDIRECT_PROMPT_INJECTION]: The skill interpolates user-provided variables (e.g.,
BASE_URL,MFA_ENDPOINT,SESSION_COOKIE) directly into shell scripts without sanitization or escaping. This creates a vulnerability surface where a malicious input could lead to arbitrary command execution on the system running the agent. - Ingestion points: Variables defined in the skill's frontmatter and workflow:
BASE_URL,MFA_ENDPOINT,POST_MFA_PAGE,SESSION_COOKIE,KNOWN_OTP, andMFA_PARAM. - Boundary markers: None. The variables are interpolated directly into backtick or double-quoted shell strings.
- Capability inventory: The skill uses
curl(network requests and file writing via-c),parallel(concurrent process execution),grep, and standard shell utilities. - Sanitization: Absent. There is no validation or escaping of the input strings before they are passed to the shell environment.
Audit Metadata