verify-login-logout
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEREMOTE_CODE_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill instructs the user to execute
npx @reticlehq/server@latest init. This command downloads and executes the latest initialization script from the vendor's official package on the NPM registry. - [EXTERNAL_DOWNLOADS]: The skill uses
curlto fetch capability documentation and LLM-specific instructions from the vendor's domains (docs.reticle.sh). It also references the vendor's GitHub repository for additional resources. - [INDIRECT_PROMPT_INJECTION]: The skill interacts with external web applications, creating a surface for indirect prompt injection.
- Ingestion points: Data enters the agent's context through browser inspection tools (
reticle_look), interaction results (reticle_act_and_wait), and assertion checks (reticle_assert) that read page text, elements, and network responses. - Boundary markers: The instructions do not define specific delimiters or provide warnings to the agent regarding potential instructions embedded within the application's UI or network data.
- Capability inventory: The skill utilizes tools to simulate user actions (
reticle_act), navigate the browser (reticle_navigate), and intercept/mock network traffic (reticle_network_mock). - Sanitization: No explicit sanitization or filtering of the content retrieved from the web application is performed before it is processed by the agent.
Audit Metadata