langfuse-integration
Pass
Audited by Gen Agent Trust Hub on Apr 9, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill installs the official langfuse and llama-index-instrumentation-langfuse packages from standard registries. It establishes connections to the official Langfuse Cloud endpoint (https://cloud.langfuse.com) for persistent trace storage.\n- [COMMAND_EXECUTION]: Local scripts utilize subprocess.run with grep to verify code instrumentation and dependency removal. These operations use static arguments and are restricted to the local project directory for verification purposes.\n- [DATA_EXFILTRATION]: The skill is designed to transmit telemetry and metadata to an external provider (Langfuse). This is the primary intended function of the skill, enabling persistent and compliant audit trails for pharmaceutical workflows as documented.\n- [PROMPT_INJECTION]: The skill processes requirement documents to extract compliance metadata. While this involves ingesting external data, the risk is negligible as the data is used for telemetry tagging and does not influence the agent's control flow or instructions.
Audit Metadata