exploring-mcp-tool-quality
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill instructions facilitate the analysis of PostHog tool quality (error rates, latency) through the PostHog platform tools such as
posthog:execute-sqlandposthog:query-mcp-tool-stats. - [SAFE]: The SQL recipes provided follow performance best practices, such as mandating time ranges to limit scan volume and using proper type casting for property strings.
- [SAFE]: All external URLs and project references (e.g.,
app.posthog.com) trace back to the verified vendor's infrastructure. - [SAFE]: The skill references local file paths for shared data models and server-side logic (
mcp_harness.py), which is consistent with a well-integrated set of internal tools and does not involve untrusted remote code execution.
Audit Metadata