mcp-aurekai

mcp-aurekai

Aurekai MCP exposes 89 native akai_\* runtime operators to any MCP-compatible host, covering the full Aurekai binary family — API gateway, artifact inspection, proof bundle export, semantic embedding, batch queuing, entity detection, compression, and more. Zero external dependencies; runs locally via npx -y @aurekai/mcp.

Category
访问服务器

README

<p align="center"> <img src="https://raw.githubusercontent.com/aurekai/aurekai/main/assets/aurekai-logo.svg" alt="Aurekai" width="520" /> </p>

@aurekai/mcp — Aurekai MCP Server

0.8.0-alpha.5 · capability-native · zero dependencies · stdio + Streamable HTTP

Exposes all 9 Aurekai capability families (111 commands) as MCP tools with full protocol-level features: tool annotations, resource pagination, named prompts, _meta proof propagation, and embedded resource outputs.

Install

npm install -g @aurekai/mcp

Usage

stdio (default — for Claude Desktop, Cursor, etc.)

// claude_desktop_config.json
{
  "mcpServers": {
    "aurekai": {
      "command": "aurekai-mcp"
    }
  }
}

Streamable HTTP (optional)

AKAI_MCP_HTTP_PORT=3100 aurekai-mcp
# POST JSON-RPC to http://127.0.0.1:3100/mcp

Protocol Surface

Feature Status
tools/list — 89 operators across 9 capability families
Tool annotations (readOnlyHint, destructiveHint, idempotentHint)
resources/list — 13 aurekai:// resource URIs
resources/read — live reads for runtime/capabilities, queue/stats, models
Resource pagination (nextCursor)
Resource subscriptions (acknowledge)
prompts/list + prompts/get — 8 named capability prompts
_meta proof propagation on tool call results
Embedded resource outputs for proof-emitting tools
logging server capability
Streamable HTTP transport (AKAI_MCP_HTTP_PORT)

Capability Families

Family Operators Examples
runtime 11 akai_api, akai_queue, akai_workflow
commerce 11 akai_gate, akai_pay, akai_ledger
intake 12 akai_transcribe, akai_ingest, akai_segment
memory 11 akai_fpq, akai_fpqx, akai_embed, akai_vec
proof 8 akai_proof, akai_canon, akai_graph, akai_hash
reason 5 akai_reason, akai_physics, akai_flow, akai_learn
wire 5 akai_tel, akai_wire, akai_moq, akai_net
publish 9 akai_brief, akai_narrate, akai_pack, akai_distribute
substrate 17 akai_capability, akai_space, akai_compress

Named Prompts

Prompt Description
turn-this-call-into-a-deliverable audio → transcribe → brief → deliverable
inspect-this-artifact-lineage Resolve full Merkle lineage for an artifact
build-a-model-memory-pack FPQ compress + roundtrip + export memory pack
compare-these-reasoning-branches Dual branch diff with recommendation
generate-client-invoice-from-usage Metering records → invoice
produce-wire-device-report PCAP → SIP event + device report
run-a-release-gate proof validate + manifest verify + SLI auto-run
make-a-client-brief-from-this-audio audio → transcript → structured client brief

Resources (aurekai:// URIs)

aurekai://runtime/capabilities · aurekai://queue/stats · aurekai://ledger/portfolio aurekai://models · aurekai://model-memory · aurekai://features/{artifact} aurekai://proof/{id} · aurekai://graph/{node}/lineage · aurekai://space/{name} aurekai://wire/{capture_id} · aurekai://project/{id} · aurekai://invoice/{id} · aurekai://cms/{entry_id}

Runtime Requirement

Tools require the akai binary on PATH (from aurekai/native-runtime) or set AKAI_BIN=/path/to/akai. Without it, tools return a clear error message — no crash.

Registry Targets

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选