Skills
skills/altinity/altinity-skills/clickhouse-profiler/Gen Agent Trust Hub

clickhouse-profiler

Pass

Audited by Gen Agent Trust Hub on May 12, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes local Python scripts (tools/pareto_cut.py and tools/synthesize_conventions.py) to perform data analysis tasks such as table usage categorization and convention synthesis. These scripts are provided within the skill package and use standard libraries.
  • [DATA_EXFILTRATION]: The skill profiles ClickHouse database metadata and query history. It enforces a strict PII discipline (Hard Rule 2), requiring query normalization and comment stripping before any data is included in the generated artifact.
  • [SAFE]: The skill implements a robust security model for database interaction, specifically mandating read-only access (Hard Rule 1) and performing existence verification (Hard Rule 8) for all identifiers to prevent injection or data fabrication.
Audit Metadata
Risk Level
SAFE
Analyzed
May 12, 2026, 07:20 PM
Security Audit — agent-trust-hub — clickhouse-profiler