secure-mcp-server

Pass

Audited by Gen Agent Trust Hub on Aug 27, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection by accepting and processing user-provided data that is subsequently used in administrative API operations.
  • Ingestion points: User-provided inputs such as groupId, assetId, version (Step 1), instanceLabel, proxyUri, uri (Step 5), and policy configuration data (Step 9).
  • Boundary markers: The skill instructions do not specify the use of delimiters or 'ignore embedded instructions' warnings when handling these inputs.
  • Capability inventory: The skill has the capability to create assets in Exchange, create and deploy API instances in API Manager, and apply security policies, which are persistent administrative changes in the environment.
  • Sanitization: There are no instructions for the agent to sanitize, escape, or validate the user-provided strings before incorporating them into the API requests.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 27, 2026, 05:23 PM
Security Audit — agent-trust-hub — secure-mcp-server