browser-auth-flow
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the
Bashtool to manage session recordings and write findings to the local filesystem within the container environment. - [DYNAMIC_EXECUTION]: The skill uses
mcp__plugin_ruflo-core_ruflo__browser_evalto execute JavaScript in the context of the target website. This is used to inspectdocument.cookieand verify security flags (Secure, HttpOnly, SameSite). While necessary for its auditing purpose, it involves running code dynamically in a browser session. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external URLs provided by the user.
- Ingestion points: Reads site content, network trajectories, and cookie data using tools like
browser_snapshot,browser_get-url, andbrowser_eval(SKILL.md). - Boundary markers: None explicitly defined in the prompt instructions to isolate untrusted web content from the agent's logic.
- Capability inventory: Access to
Bash,Write, and browser interaction tools (browser_fill,browser_type,browser_eval) provides a significant functional surface (SKILL.md). - Sanitization: Employs
aidefence_scanandaidefence_has_piito identify and redact sensitive information before it is returned to the model context, which provides a layer of defense against accidental data exposure.
Audit Metadata