azuresql-db-feedback

Pass

Audited by Gen Agent Trust Hub on Jul 31, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • Tool Execution via GitHub CLI: The skill uses the gh command-line tool to create issues on the specified repository. The process includes checks for authentication status and requires a user's clear approval before any submission occurs.
  • Indirect Prompt Injection Surface: The skill processes information from logs and user prompts to construct reports, which involves handling data that could contain unexpected instructions. 1. Ingestion points: Feedback data is sourced from user prompts and container logs. 2. Boundary markers: The skill mandates showing the user the complete report and obtaining confirmation before filing an issue. 3. Capability inventory: The agent can perform file writes and execute the gh CLI tool. 4. Sanitization: Explicit redaction rules are provided to remove sensitive data like passwords and access tokens from reports.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 31, 2026, 12:28 AM
Security Audit — agent-trust-hub — azuresql-db-feedback