No install-time abuse, credential theft chain, or unrelated exfiltration was found. However, automatic execution of backend-provided code in the host page is a high-risk AI-agent capability requiring a warning.
Static reason
No blocking static signals were detected.
Trigger
A host application initializes the SDK/widget and the connected Pluno backend sends a supported tool.call.
Impact
A compromised or malicious backend response could act with the host page's browser privileges and inspect accessible page data.
Mechanism
Remote backend-directed browser JavaScript execution plus bounded network instrumentation.
Evidence
package.jsondist/product-agent-sdk.jsdist/product-agent-widget.jsREADME.md
Network endpoints3
app.pluno.ai/api/product-agent/embed/ws/api/product-agent/embed/attachments/upload