long-polling-handler

Fail

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: HIGHPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [Indirect Prompt Injection] (HIGH): The skill's primary purpose is API integration and handling long-polling responses, which constitutes a major ingestion surface for untrusted external data.
  • Ingestion points: Processes data from third-party APIs, webhooks, and long-polling handlers (SKILL.md).
  • Capability inventory: The skill explicitly requests access to Bash, Write, Edit, and Read tools, providing a wide attack surface if the agent is manipulated.
  • Boundary markers: There are no instructions provided to the agent to distinguish between its own system prompt and instructions that may be contained within the external data it processes.
  • Sanitization: The skill lacks any defined validation or sanitization logic for the data it handles.
  • [Command Execution] (MEDIUM): The inclusion of the Bash tool in the allowed-tools list without specific constraints is a risk factor. In an API integration context, this can be leveraged to execute commands derived from malicious external payloads.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 16, 2026, 12:58 PM