oracle
Ask the oracle when you're stuck. Invoke GPT-5 Pro with a custom context and files.
What is oracle?
Ask the oracle when you're stuck. Invoke GPT-5 Pro with a custom context and files.
Overview of oracle
What it is
Oracle is a CLI and MCP server that bundles a prompt with selected files, sends that context to an AI model through an API or a signed-in browser, and stores the result as a session. It is designed for developers and coding agents that need a second-model review grounded in the actual project.
Key features
- Bundles a prompt with selected files and sends to AI model via API or browser.
- Supports multiple engines: API (OpenAI, Azure OpenAI, Anthropic, Gemini, xAI, OpenRouter, etc.), Browser (ChatGPT, Gemini via Chrome automation), and Render (preview without sending).
- File selection with
--fileaccepts files, directories, globs, and exclusions. - Sessions stored locally under
~/.oracle/sessionsfor background runs and replay. - Multi-model runs with
--modelsfor API panel, recording per-model usage, cost, output. - Agent integrations via MCP server (
oracle-mcp) or Oracle skill.
Use cases
- Reviewing code or project files with a second AI model.
- Auditing specific files or directories for issues.
- Getting a second opinion on code quality, security, or release risks.
- Automating code review in CI or agent workflows.
Getting started
Install with Homebrew on macOS/Linux: bash brew install steipete/tap/oracle Or install the npm package globally: bash npm install -g @steipete/oracle Oracle requires Node.js 24 or newer. To try without installing: bash npx -y @steipete/oracle --help For a quick preview of the bundle without sending to a model: bash oracle --render -p 'Review the package metadata for release risks' --file package.json When an engine is configured, remove --render to request an answer.
License and pricing
The license is MIT. The pricing type is open source.
Based on the project's public README and repository metadata.