turning-engineering-analytics-into-insights

Pass

Audited by Gen Agent Trust Hub on Sep 6, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides structured guidance for querying PostHog data warehouse tables using standard HogQL. It uses legitimate MCP tools such as execute-sql, insight-create, and dashboard-create for their intended purposes.
  • [SAFE]: Data access is scoped to the team's own GitHub warehouse tables (e.g., github_pull_requests, github_workflow_runs) and provisioned views (e.g., engineering_analytics_job_costs). No unauthorized data access or sensitive file exposure was detected.
  • [SAFE]: The HogQL recipes follow best practices for data processing in PostHog, including timestamp parsing, bot filtering, and team-level aggregation to prevent per-developer surveillance.
  • [SAFE]: All external references point to internal documentation (references/hogql-recipes.md) and standard PostHog platform features without remote code execution or suspicious downloads.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 6, 2026, 07:58 AM
Security Audit — agent-trust-hub — turning-engineering-analytics-into-insights