oracle: what it is, what problem it solves & why it's gaining traction

What it solves

Oracle is a CLI tool designed to bundle prompts and local files into a single context window for AI models. It eliminates the manual effort of copying and pasting large sets of of files into an AI chat interface, allowing users to provide real-world context from their codebase or documents to get more accurate answers.

How it works

Oracle operates through two primary engines: an API engine that connects directly to providers like OpenAI, Google (Gemini), and Anthropic, and an experimental browser engine that automates a Chrome browser to interact with ChatGPT or Gemini without requiring API keys. It supports multi-model runs, where a single prompt can be sent to multiple models simultaneously to compare results. It also features session management, allowing users to follow up on existing conversations (via --followup) and replay previous sessions.

Who it’s for

It is aimed at developers and power users who frequently use LLMs for code review, architecture planning, or bug reporting and want a streamlined way to integrate local file context into their AI workflows.

Highlights

• Flexible Engine Choice: Supports both direct API access and browser automation for those without API keys.
• Multi-Model Advisory: Ability to query multiple models in one run to cross-check answers.
• Context Bundling: Uses globs and excludes to precisely attach the necessary files and directories.
• MCP Integration: Provides an MCP server (oracle-mcp) for integration with AI agents like Claude Code and Cursor.
• Session Lineage: Tracks and allows continuation of conversations through session IDs and response IDs.