azure-messaging-webpubsubservice-py

Warn

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • Unverifiable Dependencies & Remote Code Execution (LOW): The skill documentation instructs the installation of official Microsoft Azure SDK packages from a trusted repository for real-time messaging capabilities.\n
  • Evidence: Installation commands for azure-messaging-webpubsubservice and azure-messaging-webpubsubclient provided in SKILL.md.\n
  • Trust Scope: Downgraded to LOW as the organization 'Azure' is a verified trusted source in the provided framework.\n- Indirect Prompt Injection (MEDIUM): The skill creates an attack surface by allowing the agent to ingest data from external WebSocket connections and providing the capability to send messages to various targets.\n
  • Ingestion points: WebSocket event handlers such as client.on("server-message") and client.on("group-message") in SKILL.md ingest untrusted external data.\n
  • Boundary markers: There are no boundary markers or instructions to ignore embedded commands within the message payloads demonstrated in the provided code.\n
  • Capability inventory: The skill exposes functions to broadcast or target messages (send_to_all, send_to_user, send_to_group) and manage user groups, which could be leveraged if the agent is manipulated by untrusted input.\n
  • Sanitization: The example code does not include any sanitization or validation of the message content before processing or re-transmitting.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 16, 2026, 09:51 AM