exploring-llm-clusters

Pass

Audited by Gen Agent Trust Hub on Apr 7, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides instructions and tools for analyzing LLM clusters and traces specifically for the PostHog platform. All operations (SQL queries, trace inspections) are scoped to legitimate analytical workflows.
  • [COMMAND_EXECUTION]: The skill includes a helper script scripts/print_clusters.py. Analysis of the script shows it is a benign utility for parsing and summarizing JSON cluster data from local files. It does not perform network operations or execute arbitrary system commands.
  • [EXTERNAL_DOWNLOADS]: No external downloads or remote script executions are performed. All functionality relies on the provided script and the posthog toolset.
  • [DATA_EXFILTRATION]: The skill processes analytical data (trace IDs, costs, latencies) within the authorized PostHog environment. No unauthorized network requests or data transmission to third-party domains were found.
  • [PROMPT_INJECTION]: The instructions do not contain patterns that attempt to bypass safety filters or override the agent's core instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 7, 2026, 09:03 AM