TeaQL Agent Kit
A kit that gives an existing coding agent a model-first workflow for modeling a business domain, generating an application from that model, and verifying the result.
TeaQL Agent Kit gives a coding agent such as Claude Code, Codex, Cursor, or GitHub Copilot a structured, model-first workflow built around TeaQL: model the business domain, evaluate and repair that model, generate an application from it, and verify the outcome, without the developer first learning TeaQL's runtime, generator, or toolchain.
A user starts the workflow with a single prompt describing their own requirement, and the agent installs or updates TeaQL's clients and language build tools, downloads dependencies, and calls TeaQL's services for model evaluation, generation, and API guidance, which requires network access and sends the described model content to that service. The kit's query API lets application code express what data to load, such as filters and related objects, as typed, reviewable queries instead of hand-written query-building and data-access code repeated in every method.
It is written in Python and released under the MIT license.