api-mock-server

Pass

Audited by Gen Agent Trust Hub on Mar 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill requires running a persistent Node.js server process (scripts/server.js) on the host machine to provide mock API functionality.- [COMMAND_EXECUTION]: The server exposes a management endpoint (POST /_api/shutdown) that triggers process.exit(0), allowing the server to be terminated programmatically via network request.- [PROMPT_INJECTION]: The skill presents an attack surface for indirect prompt injection because its primary purpose is to serve data that an AI agent will consume and act upon.
  • Ingestion points: Data enters the system via the /_api/routes management API and through incoming HTTP request headers, query parameters, and bodies that are evaluated against mock conditions.
  • Boundary markers: The server does not implement or enforce boundary markers or instructions for agents to ignore instructions embedded in the mock response data.
  • Capability inventory: The server has the ability to write configuration to the local file system (scripts/data/routes.json) and control its own lifecycle (shutdown). It does not have general shell execution capabilities.
  • Sanitization: There is no sanitization or validation performed on user-configured mock responses or the regex patterns used in the condition matcher, which could be exploited for local resource exhaustion (ReDoS).
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 13, 2026, 08:06 AM