Topic

Cursor

1 min read

Cursor MCP Config vs an OpenAPI-Generated MCP Server

Short answer: A hand-written MCP entry in Cursor's config is plumbing you maintain; a generated server is derived from the contract. Both expose tools to the agent, but

Read article
1 min read

Why Cursor Hallucinates API Calls Without a Spec

Short answer: An agent hallucinates an API call because it is predicting the next token, not reading your server. Without a contract, it fills in the endpoint, parameters, and body

Read article
1 min read

API Testing in Cursor vs a Contract-Verified Workflow

Short answer: An editor can write a test, but it cannot prove the response conforms to the API. A contract-verified workflow checks each response against the schema and flags the

Read article
1 min read

Local-First API Mocking for Cursor Agents

Short answer: Give the agent a local mock generated from the spec, not a live backend. It is deterministic, contract-shaped, and returns the edge cases on demand, with no account

Read article
1 min read

Cursor and Powerduck: Why They Are Better Together

Short answer: The two tools sit at different layers and compose cleanly. Cursor writes the code and drives the agent; Powerduck defines the contract, generates the typed tools and mock,

Read article