MCP server for AI agent integration with PortZero - enables agents to manage local development apps, inspect traffic, replay requests, and access inferred API schemas
PortZero MCP server has 7 well-named tools with clear action verbs (list_, get_, replay_, restart_). All tools have descriptions (avg 95 chars, within 10-1024 range). Input schemas are present and properly typed with JSON Schema. However, parameter descriptions are sparse or missing in several tools (e.g., get_recent_requests has 6 params but only 4 have descriptions; replay_request's 'headers' param lacks detail on merge semantics). Output schemas are not documented, LLMs cannot predict response structure. Error handling is minimal; no recovery guidance or actionable error messages visible. Tool composition is sound (single responsibility), but missing pagination guidance for list_apps and get_recent_requests despite potentially large result sets.
Get the last N log lines for a specific app
Get the inferred API schema for an app, built from observed traffic. Shows endpoints, methods, parameters, and status codes.
Get recent HTTP requests captured by the proxy, optionally filtered by app, status, or method
Get the full request and response details for a specific request ID, including headers and body
List all running apps with status, port, and URL
Re-send a previously captured request with optional overrides to method, URL, headers, or body
Restart a running or crashed app
Output schemas not documented. LLMs cannot predict response structure for any tool, forcing them to guess at field names and types for downstream chaining.
get_recent_requests has 6 parameters but only 4 have descriptions. 'limit' and 'status_range' lack explanation of valid values and constraints.
replay_request 'headers' parameter description does not explain merge semantics (are original headers preserved? overwritten? merged?). LLMs will guess incorrectly.
No pagination guidance for list_apps and get_recent_requests. If result sets grow large, LLMs will receive truncated data without knowing more exists.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | C | 60 | 2026-07-28+ | v2 |
No error handling guidance visible. Tools do not document what errors are retryable, user-fixable, or fatal, or how to recover.