exploring-mcp-tool-quality

Pass

Audited by Gen Agent Trust Hub on Sep 11, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes event data from PostHog's database, which is an external ingestion point. (1) Ingestion points: Data is read from the events table via posthog:execute-sql and other analytics tools in SKILL.md. (2) Boundary markers: Not explicitly defined in the skill instructions. (3) Capability inventory: The skill can execute SQL queries and query specific analytics endpoints. (4) Sanitization: Instructions recommend casting property strings to specific types (toBool, toFloat), providing a layer of data validation.
  • [COMMAND_EXECUTION]: The skill uses posthog:execute-sql to perform analytical queries on the events table. The provided SQL recipes include safety measures such as time-range filters (timestamp >= now()
  • INTERVAL 30 DAY), row limits (LIMIT 20), and volume floors (HAVING total_calls >= 20) to prevent performance issues.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 11, 2026, 10:55 AM
Security Audit — agent-trust-hub — exploring-mcp-tool-quality