# Hister — Self-Hosted Search Engine (MCP-ready)

> Hister is a privacy-focused self-hosted search engine that full-text indexes visited sites locally, with optional semantic search and MCP agent access.

## Install

Merge the JSON below into your `.mcp.json`:

## Quick Use

```bash
git clone https://github.com/asciimoo/hister.git
cd hister
./manage.sh build
# dev webui (HMR + backend rebuild)
npm run serve:app
```

## Intro

Hister is a privacy-focused self-hosted search engine that full-text indexes visited sites locally, with optional semantic search and MCP agent access.

**Best for:** People who want local search over browsing/indexed content without remote engines

**Works with:** Go + npm for dev, self-hosted deployment, optional MCP integrations

**Setup time:** 10-25 minutes

### Key facts (verified)

- GitHub: 906 stars · 48 forks · pushed 2026-05-13.
- License: AGPL-3.0 · owner avatar + repo URL verified via GitHub API.
- README-verified entrypoint: `./manage.sh build`.

## Main

- Decide your indexing scope first: local browsing history vs. explicit crawl (README mentions both headless browser crawling and traditional crawler modes).

- If you enable AI features, start with semantic search, then consider MCP-based agent connections only after you’re happy with the local index quality.

- For teams, use the multi-user hosting mode and keep it inside your trusted network to preserve the privacy value proposition.

### Source-backed notes

- README describes local full-text indexing for visited websites and a query language for advanced search.
- README lists optional AI enhancements: semantic search and connecting agents via MCP.
- README Development section says you can build via `./manage.sh build` and run a dev server via `npm run serve:app`.

### FAQ

- **Is it private by default?**: README emphasizes local indexing to avoid remote search engines for day-to-day queries.
- **Do I need AI features?**: No — full-text local search works without them; semantic search/MCP are optional add-ons per README.
- **What do I need to build it?**: README lists requirements as ‘latest Go and NPM’ and a `./manage.sh build` flow.

## Source & Thanks

> Source: https://github.com/asciimoo/hister
> License: AGPL-3.0
> GitHub stars: 906 · forks: 48

---

<!-- ZH -->

## Quick Use

```bash
git clone https://github.com/asciimoo/hister.git
cd hister
./manage.sh build
# dev webui (HMR + backend rebuild)
npm run serve:app
```

## Intro

Hister 是隐私优先的自托管搜索引擎：对访问过的网页做本地全文索引，支持查询语言与本地知识库索引，并可用 crawler 扩展范围，减少对远程搜索引擎的依赖；可选语义搜索，并可通过 MCP 接入 agent。

**Best for:** 想在本地做网页/知识库搜索、尽量不依赖远程搜索引擎的人

**Works with:** Go + npm 开发依赖；自托管部署；可选 MCP 集成

**Setup time:** 10-25 minutes

### Key facts (verified)

- GitHub：906 stars · 48 forks；最近更新 2026-05-13。
- 许可证：AGPL-3.0；作者头像与仓库链接均已通过 GitHub API 复核。
- README 中核对过的入口命令：`./manage.sh build`。

## Main

- 先决定索引范围：是仅索引访问过的网站，还是启用爬虫快速扩展（README 同时提到 headless browser 与传统 crawler）。

- 开启 AI 功能时建议先上语义搜索；当本地索引质量稳定后，再考虑通过 MCP 把能力暴露给 agent。

- 团队使用时可按 README 的 multi-user 思路在内网自托管，才能发挥隐私优势。

### Source-backed notes

- README 说明它会对访问过的网站做本地全文索引，并提供查询语言用于高级检索。
- README 写到可选 AI 增强：语义搜索，以及通过 MCP 连接 agent。
- README 的 Development 区域给出 `./manage.sh build` 的构建方式，以及 `npm run serve:app` 的开发模式。

### FAQ

- **默认就隐私吗？**：README 强调本地索引/本地搜索，减少依赖远程搜索引擎，这是其主要价值点。
- **一定要用 AI 功能吗？**：不必。全文检索本身即可用；语义搜索与 MCP 连接是可选项。
- **构建需要什么环境？**：README 写的是最新 Go 与 NPM，并提供 `./manage.sh build` 的构建流程。

## Source & Thanks

> Source: https://github.com/asciimoo/hister
> License: AGPL-3.0
> GitHub stars: 906 · forks: 48


---
Source: https://tokrepo.com/en/workflows/hister-self-hosted-search-engine-mcp-ready
Author: MCP Hub