neo4j-security-skill
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFECREDENTIALS_UNSAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [CREDENTIALS_UNSAFE]: The skill documents the Cypher command 'SHOW USERS WITH AUTH AS COMMANDS', which exports user credentials and authentication provider configurations. This represents a potential data exposure risk if the agent is manipulated into leaking the command's output.
- [INDIRECT_PROMPT_INJECTION]: The skill processes complex security rules (PBAC and ABAC) that may be based on untrusted external data like OIDC claims or user tags, creating a surface for logic-based injection attacks. The skill mitigates this risk by requiring a mandatory confirmation gate ('MCP Write Gate') for all security-sensitive commands.
Audit Metadata