funnel-analysis-builder

Fail

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: HIGHCOMMAND_EXECUTION
Full Analysis
  • Indirect Prompt Injection (HIGH): The skill is designed to process external data sources (SQL, data visualization inputs) and has access to powerful tools including Bash, Write, and Edit. This creates a critical vulnerability surface where malicious instructions embedded in data files could hijack the agent's high-privilege capabilities.
  • Ingestion points: Processes data for 'funnel analysis builder' and 'sql queries' mentioned in SKILL.md.
  • Boundary markers: None defined in the skill configuration to separate instructions from data.
  • Capability inventory: Access to Bash, Write, and Edit allows for system-level changes and arbitrary command execution.
  • Sanitization: No sanitization or validation logic is defined to prevent embedded instructions from being executed via the Bash tool.
  • Command Execution (HIGH): The explicit request for the Bash tool enables the execution of any shell command. Without strict constraints or a trusted author, this capability can be abused to perform unauthorized actions on the host system if the agent is manipulated by adversarial input.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 17, 2026, 12:00 AM