steipete/oracle

Ask the oracle when you're stuck. Invoke GPT-5 Pro with a custom context and files.

What it solves

Oracle is designed for developers and coding agents who need a second-model review of their project. It solves the problem of manually copying and pasting files into an AI model to get a review or audit, by automating the bundling of a prompt and selected project files into a context window.

How it works

Oracle operates as a CLI and an MCP server. It allows users to select files, directories, or globs (with exclusions) to create a a context bundle. This bundle is then sent to an AI model via an API (supporting OpenAI, Azure OpenAI, Anthropic, Gemini, xAI, and OpenRouter) or through a signed-in browser session (using Chrome automation for ChatGPT and Gemini).

Who it’s for

It is built for developers and coding agents that require grounded, project-specific AI reviews and audits.

Highlights

  • Flexible Context Control: Supports files, directories, and globs with exclusions to precisely define the context.
  • Multiple Engine Modes: Offers API mode for automation, Browser mode for using existing signed-in sessions, and Render mode for local inspection of the prompt bundle.
  • Session Management: Stores runs in local sessions, allowing users to reattach to long responses, replay answers, and perform follow-up queries.
  • Multi-Model Support: Can run a single prompt across multiple models simultaneously to compare outputs and track costs.
  • MCP Integration: Includes an oracle-mcp stdio server for integration with MCP clients like Claude Code, Codex, and Cursor.

Related

  • Project
    oracle/mcp

    Reference implementations of Model Context Protocol (MCP) servers that expose Oracle Cloud tools (via OCI SDK, OCI CLI, Java, JavaScript, etc.) in a language‑agnostic way. Includes polyglot servers, shared authentication, container support, and integration with MCP clients like Cline, Cursor, and MCPHost. Intended for prototyping and learning, not production.

  • Project
    apify/mcpc

    mcpc is a universal command‑line client for the Model Context Protocol (MCP). It maps every MCP operation—tools, prompts, resources, tasks, skills, OAuth authentication, and even experimental x402 payments—into intuitive shell commands. Designed for AI agents and developers, it lets you inspect, debug, and script MCP workflows using standard Unix tools (e.g., jq, xargs). Features include persistent named sessions, progressive tool discovery via `grep`, JSON‑only output mode, secure OAuth 2.1 storage (OS keychain or protected file), AI sandboxing, and lightweight cross‑platform support.

  • Project
    microsoft/mcp

    The Model Context Protocol (MCP) is an open standard that enables AI applications to connect to data sources and tools consistently. This repository provides the core infrastructure and a collection of official Microsoft servers to integrate services like Azure, GitHub, and Microsoft 365 with AI agents.

  • Project
    1mcp-app/agent

    1MCP is a Node.js runtime that aggregates multiple MCP servers into one process and offers a CLI mode for AI coding agents (Codex, Claude, Cursor). Agents discover tools step‑by‑step (instructions → inspect → run), reducing configuration and tool sprawl while sharing auth, presets, and per‑repo templates.

  • Project
    googleapis/gcloud-mcp

    A collection of Model Context Protocol (MCP) servers that enable AI assistants to interact with Google Cloud via the gcloud CLI and various cloud APIs using natural language.