Ross ROSS = Recommend OSS · open-source software intelligence for agents

steipete/oracle

Ask the oracle when you're stuck. Invoke GPT-5 Pro with a custom context and files. observed · 2026-08-28

github.com/steipete/oracle · homepage · TypeScript · MIT (permissive) observed · 2026-08-28

Health v2 · maintenance only

82/100

  • Activity 99
  • Release rhythm 97
  • Longevity 20
How is this computed?

round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10) — computed 2026-09-02. Adoption (stars, forks) is never an input.

  • gap_med: 2.0
  • age_days: 292
  • days_rel: 20
  • days_push: 10
  • n_releases_24m: 49

Full methodology

Adoption not part of the score

3648 stars · 377 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

Oracle is a TypeScript CLI and MCP server that bundles a prompt with selected project files and sends that context to frontier AI models (GPT-5.x Pro, Gemini, Claude, OpenRouter) via provider APIs or a signed-in Chrome browser session. It stores each consult as a replayable session and supports multi-model cross-checks in a single run.

Use cases

  • get a second-model review of my code with real repository context
  • ask GPT-5 Pro a question with files attached from the terminal
  • cross-check assumptions across multiple AI models in one run
  • use my ChatGPT Plus subscription from the CLI without an API key
  • expose an AI consult tool to coding agents via MCP
  • assemble a prompt bundle to paste into ChatGPT manually
  • follow up on a previous AI consult with new context

When to choose

  • you want pro-tier model answers grounded in actual project files
  • you need multi-model comparison or cost/token tracking in one command
  • you want to give coding agents a second-opinion tool via MCP
  • you prefer browser mode to avoid paying API rates

When to avoid

  • you need a fully managed hosted service or GUI rather than a CLI
  • you cannot run Node.js 24+ or install Chrome for browser mode
  • you want fully offline inference with local models

Facets

cli-tool · maturity active

cli llm-inference agent-framework mcp developer-tools developer-tools large-language-models windows cli second-opinion multi-model chatgpt-browser-automation code-review prompt-bundling openai anthropic gemini openrouter ai-agents command-line macos linux nodejs

5 sources

Member repositories

RepositoryRoleHealth v2
steipete/oraclemain82

For agents

markdown · JSON · MCP: product_card(name="steipete/oracle")

Data as of 2026-08-30T08:39:29.467469+00:00 · Report a problem