Help us improve
Share bugs, ideas, or general feedback.
Share bugs, ideas, or general feedback.
Share bugs, ideas, or general feedback.
AI-powered test generation for Claude Code — generate, init, and migrate test suites via slash commands.
npx claudepluginhub bri-stevenski/oracle-test-ai-agent --plugin oracleDiagnose a flaky test and propose a deterministic fix from code, CI logs, or a failure description.
Recommend the right testing framework and folder structure for a given need.
Review existing test code for brittleness, anti-patterns, and coverage gaps.
Generate production-ready test code from a natural-language requirement (Playwright, Vitest, Pytest, k6).
Diagnose flaky tests by analyzing test code, CI logs, and failure patterns, then propose a deterministic fix. Use when the user says "this test is flaky", "intermittent failure", "passes locally fails in CI", "only fails sometimes", or pastes a CI log of a failing test. NOT for tests that fail consistently — those are bugs, not flakes.
Recommend the right testing framework (Playwright, Vitest, Pytest, k6, contract-testing tools) for a given testing need, and propose a folder structure. Use when the user asks "what framework should I use", "Playwright or Cypress", "how should I structure my tests", or "what kind of test is right for X". Recommends only — does not write the tests themselves.
Scaffolds a new test suite for a chosen framework using Oracle's init tool.
Migrates a harness-scaffolded test suite to Oracle's layout with an explicit confirm-before-apply flow.
Generate production-ready test code from natural-language requirements across Playwright (E2E), Vitest (JS/TS unit and component), Pytest (Python unit and API), and k6 (performance). Use when the user wants to *create* new tests — phrases like "write a test for...", "generate an E2E test", "I need a unit test that covers X", or "scaffold tests for this module".
> Add a new testing framework to Oracle's registry end-to-end. Confirms
> Generate a framework-appropriate test from a natural-language
> Move a generated test from `tests/generated/` into the committed test
> Configure the Harness Engineering guardrails in a new Oracle
Executes bash commands
Hook triggers when Bash tool is used
Modifies files
Hook triggers on file write and edit operations
Uses power tools
Uses Bash, Write, or Edit tools
Share bugs, ideas, or general feedback.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge.
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge.
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Complete collection of battle-tested Claude Code configs from an Anthropic hackathon winner - agents, skills, hooks, and rules evolved over 10+ months of intensive daily use
Efficient skill management system with progressive discovery — 410+ production-ready skills across 33+ domains
Comprehensive .NET development skills for modern C#, ASP.NET, MAUI, Blazor, Aspire, EF Core, Native AOT, testing, security, performance optimization, CI/CD, and cloud-native applications
Unity Development Toolkit - Expert agents for scripting/refactoring/optimization, script templates, and Agent Skills for Unity C# development
Complete creative writing suite with 10 specialized agents covering the full writing process: research gathering, character development, story architecture, world-building, dialogue coaching, editing/review, outlining, content strategy, believability auditing, and prose style/voice analysis. Includes genre-specific guides, templates, and quality checklists.
Share bugs, ideas, or general feedback.
Oracle is an AI-powered test automation agent that transforms natural language requirements into high-quality, framework-aware test code.
pipx install git+https://github.com/bri-stevenski/oracle-test-ai-agent@v0.2.0
This installs the oracle binary globally without polluting your project
virtualenv. Upgrade by re-running with @<new-tag>.
git clone https://github.com/bri-stevenski/oracle-test-ai-agent.git
cd oracle-test-ai-agent
pip install -e .
Oracle ships as a Claude Code plugin with four MVP personas — oracle-test-author,
oracle-test-reviewer, oracle-framework-advisor, and oracle-flake-hunter —
plus the harness MCP server. Add it as a local marketplace:
/plugin marketplace add https://github.com/bri-stevenski/oracle-test-ai-agent
/plugin install oracle@oracle
Or, if you've already cloned the repo:
/plugin marketplace add /path/to/oracle-test-ai-agent
/plugin install oracle@oracle
Note: the plugin shells out to the
oracleCLI for code generation, so install the Python package first (above) and setOPENAI_API_KEYbefore invoking the test-author persona.
To generate a test, simply provide a prompt. Oracle will classify the intent, recommend a framework, and generate the code.
oracle generate "Create a playwright test for the login flow on https://example.com"
If you want to see Oracle's reasoning without calling the LLM (and without
needing an API key), use the --recommend-only flag:
oracle generate "I need a load test for my checkout API" --recommend-only
oracle version
Oracle defaults to Claude (Anthropic). To use the generation features, set your API key:
export ANTHROPIC_API_KEY='your-anthropic-api-key'
Other providers (gemini, openai, mock) are available via the
ORACLE_LLM_PROVIDER environment variable. See
docs/wiki/LLM-Providers-and-Configuration.md
for the full provider matrix and required env vars.
Oracle follows a modular pipeline: User Prompt → Classifier → Recommender → Orchestrator → LLM → Generated Test
Generated by Oracle — The future of autonomous test engineering.