The flagship project in the JC Meteor portfolio.
MeteorTest is an automation testing platform for managing projects, importing suites, scheduling local executors, collecting reports, and using AI to assist with project import, suite operations, task creation, report inspection, and failure analysis.
The automation repository provides meteortest.yml, declaring suites, commands, required tools, and report behavior.
MeteorTest reads the contract and registers the project suites, either through the Web console or through AI-assisted project operations.
A user or AI assistant can create a task that binds the project, suite, environment, build artifact, and execution metadata into one run request.
The Python Local Agent claims the task and runs the declared pytest, Appium, or Locust command on the local machine.
Logs, status, Allure artifacts, screenshots, and execution summaries are written back to the platform.
MeteorTest uses platform context to help import suites, create tasks, inspect reports, summarize failures, and suggest the next action.
Automation testing work often starts inside one repository, then spreads across products, environments, devices, reports, and local machines. MeteorTest is an attempt to make that execution loop visible and repeatable without moving every test into the platform itself.
The hard part is not only running pytest or Appium. The hard part is knowing which project, suite, app artifact, environment, executor, log, report, failure summary, and AI operation belong to the same run.
MeteorTest keeps the platform as the control plane and lets the Local Agent handle execution. Test repositories own their code and expose a meteortest.yml contract; the platform imports that contract and schedules tasks against it.
Owns project metadata, suite import, task state, reports, executor visibility, settings, and AI-assisted operation surfaces.
Claims queued tasks, prepares artifacts, runs suite commands, captures logs, and writes reports back to the platform.
Owns test code, fixtures, app-specific configuration, report output, and the meteortest.yml integration contract.
A MeteorTest Local Agent task executed iOS-Automation-Framework `api_smoke` against the deterministic local mock API and collected task-specific logs plus Allure result artifacts.
The public website still uses a browser-side mock demo, and the live Web preview keeps execution private. Public connected execution is a later design topic after authentication, data isolation, and executor safety are designed.
These screenshots were captured from a local MeteorTest Web preview using sanitized mock data. They show the current console structure without exposing private Supabase data, local paths, credentials, or real test accounts.
The dashboard shows task status, project onboarding, executor health, AI-assisted entry points, and the execution loop in a populated console layout.
The report page shows task outcomes, failure context, report summaries, and AI analysis surfaces with mock execution data.
The AI page presents conversations, executable quick templates, task-oriented prompts, and the assistant workspace as an operations surface rather than a simple chat box.
This is not a public connected demo. MeteorTest Web is now online as a public preview, but Local Agent execution remains private.
The current website demo is an interactive browser walkthrough that explains the project import, task preparation, Local Agent handoff, report context, and AI follow-up flow. The separate MeteorTest Web preview is online at meteortest.jcmeteor.com for console surface validation. Local Agent execution remains private, and public connected execution is still deferred.
Open interactive demo Open Web preview