skills/awslabs/mcp/aurora dsql/Gen Agent Trust Hub

aurora dsql

Pass

Audited by Gen Agent Trust Hub on Feb 27, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill includes several Bash scripts for cluster management and database connectivity. These tools use standard AWS CLI-like functionality to manage infrastructure. Evidence: Scripts located in the scripts/ directory, including create-cluster.sh, delete-cluster.sh, list-clusters.sh, cluster-info.sh, and psql-connect.sh.\n- [EXTERNAL_DOWNLOADS]: The documentation references official AWS resources and sample repositories for implementation guidance. Evidence: Links to docs.aws.amazon.com and github.com/aws-samples/aurora-dsql-samples.\n- [PROMPT_INJECTION]: The skill identifies a surface for indirect prompt injection (SQL injection) due to the nature of direct SQL query execution without built-in parameterization. Evidence Chain: 1. Ingestion points: readonly_query and transact tools in SKILL.md. 2. Boundary markers: Absent in the tool definitions; the skill relies on instructional warnings. 3. Capability inventory: Tools allow execution of DDL and DML; scripts allow cluster creation and deletion. 4. Sanitization: Absent in the tools; the documentation mandates manual sanitization using allowlists and regex as a critical developer responsibility.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 27, 2026, 03:36 PM