langfuse-trace-view

Pass

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: LOWPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION] (LOW): Indirect Prompt Injection surface detected in the trace viewing functionality.
  • Ingestion points: The script scripts/langfuse-trace-view.ts fetches trace data including 'input' and 'output' fields from the Langfuse API.
  • Boundary markers: None. External trace content is printed directly to the console output without delimiters or instructions to ignore embedded commands.
  • Capability inventory: The skill is limited to fetching and displaying information; it does not have write access or the ability to execute commands based on the trace content.
  • Sanitization: The script performs basic truncation of long JSON strings but does not sanitize the content for potential malicious instructions.
  • [COMMAND_EXECUTION] (LOW): The skill executes a local TypeScript script using npx tsx. While this is a common execution pattern for Node.js-based AI agent skills, it relies on the execution of code within the local environment.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 16, 2026, 07:11 AM