Skip to content

Jiro Search 0.1.0 (MVP)

Choose a tag to compare

View all tags
@DevAnimecx DevAnimecx released this 25 Aug 19:42
· 56 commits to main since this release
v0.1.0
c4d1dc9

Jiro Search 0.1.0 - Initial Public MVP

Local-first, AI-native web search & scraping API - a self-hosted SerpAPI alternative with an MCP server, agentic research, and built-in legal compliance. MIT licensed, free forever.

Highlights

  • 9 search engines (Google, Bing, DuckDuckGo, Brave, YouTube, Amazon, eBay, Yandex, Baidu) with automatic fallback
  • Universal scraper: URL -> markdown/text/html/JSON, readability + metadata extraction
  • Agentic research (/ai/search, /ai/agent) with citations + SSE streaming
  • MCP server (stdio / Streamable HTTP / SSE) for Claude Desktop, Cursor, Continue.dev, Zed, Cline
  • AI-agent tooling: OpenAI/Anthropic/Gemini function-calling + LangChain/LlamaIndex wrappers
  • Compliance layer: robots.txt parsing, ToS acknowledgments (persisted to SQLite), audit logging
  • BYOK: proxies (BrightData/Oxylabs/ScraperAPI/ZenRows/Smartproxy), CAPTCHA (2Captcha/CapSolver), LLM keys
  • Team features: hashed API keys, roles/scopes, per-key rate limits, JWT, /usage + Prometheus /metrics
  • Ops: Docker + Helm, async jobs with HMAC-signed webhooks

Get started

pip install jiro-search && jiro serve -> http://localhost:8000/docs

Open-core: free MIT community edition; Jiro Cloud + Enterprise (BSL-1.0) planned.

Assets 2
Loading