langfuse-hello-world

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill recommends the installation of standard observability and tracing packages from official registries.
  • Evidence: The Error Handling section and code comments suggest installing @langfuse/tracing, @langfuse/otel, and @opentelemetry/sdk-node via npm, as well as the langfuse Python package.
  • [INDIRECT_PROMPT_INJECTION]: The skill demonstrates how to trace and capture user-supplied input, which creates a potential surface for indirect prompt injection if the data contains malicious instructions targeting the LLM or observability platform.
  • Ingestion points: Multiple functions in SKILL.md (e.g., processQuery, generateAnswer, process_query) accept and trace arbitrary string inputs.
  • Boundary markers: The provided boilerplate code does not utilize specific delimiters or boundary markers to isolate untrusted user data from the tracing context.
  • Capability inventory: The skill uses network capabilities to transmit the captured trace data to the Langfuse backend.
  • Sanitization: No input validation or sanitization logic is included in the examples for the external content being processed.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 03:47 AM
Security Audit — agent-trust-hub — langfuse-hello-world