@ainecto/mcp
Ainecto MCP connector and CLI that proxies MCP protocol to Ainecto's remote endpoint, enabling tool listing, project management, and task operations through natural language.
README
@ainecto/mcp
Ainecto MCP connector and first-party CLI.
This package exposes two entry points:
npx -y @ainecto/mcporainecto mcp: raw stdio-to-HTTP JSON-RPC proxy for Ainecto MCP.ainecto <command>: human CLI for auth and low-level tool calls.
Phase 1 keeps the connector as a raw proxy. File, stdin, and inline JSON parsing are only available in first-party CLI commands.
Local Setup
npm install
npm run typecheck
npm test
npm run build
npm link
Commands
ainecto auth login --env dev
ainecto auth status --env dev
ainecto auth logout --env dev
ainecto tools list --env dev
ainecto tools call mcp__ainecto__list_projects --env dev --json
ainecto tools call mcp__ainecto__erd_apply_changes -f changes.json --json
cat payload.json | ainecto tools call mcp__ainecto__erd_apply_changes --json
ainecto projects list --env dev
ainecto task list-tasks --env dev --document-uuid <documentUuid> --json
ainecto erd apply-changes --env dev -f erd-operations.json --yes
Endpoint resolution priority:
--endpoint <url>AINECTO_MCP_ENDPOINT--env dev- production default,
https://ainecto.com/mcp
Endpoint URLs must use https:. Plain http: is accepted only for localhost loopback targets such as 127.0.0.1. When AINECTO_TOKEN is set, the CLI only sends it to the default prod/dev endpoints unless AINECTO_ALLOW_CUSTOM_ENDPOINT_TOKEN=1 is set for an explicitly trusted custom endpoint.
Connector Mode
npx -y @ainecto/mcp
ainecto mcp --env dev
The connector proxies initialize, tools/list, and tools/call to the resolved /mcp endpoint. It does not rewrite remote schemas or interpret local file references.
MCP Client Installation
Use the direct package configuration until the server is published to the official MCP Registry.
{
"mcpServers": {
"ainecto": {
"command": "npx",
"args": ["-y", "@ainecto/mcp"]
}
}
}
For a non-production endpoint, pass CLI flags through the package args:
{
"mcpServers": {
"ainecto-dev": {
"command": "npx",
"args": ["-y", "@ainecto/mcp", "--env", "dev"]
}
}
}
The planned official Registry server name is io.github.codelives/ainecto, backed by the public npm package @ainecto/mcp. Registry publication requires the npm package version referenced by server.json to include a matching mcpName field in package.json.
Catalog Sync
sync:tools --check is intended for publish-time live drift checks. It fails fast when AINECTO_CATALOG_SYNC_TOKEN is missing, so automatic publish cannot silently fall back to fixtures.
For local development, sync:tools can also use credentials from ainecto auth login --env <env>.
AINECTO_CATALOG_SYNC_TOKEN=... npm run sync:tools -- --env prod --check
AINECTO_CATALOG_SYNC_TOKEN=... npm run sync:tools -- --env dev --check
The checked-in generated catalogs are deterministic output from tools/list. The prod and dev catalogs are live-synced snapshots from https://ainecto.com/mcp and https://dev.ainecto.com/mcp. Presentation metadata lives separately in src/core/catalog/enrichments.ts.
ainecto tools catalog prints the local generated catalog. Rerun authenticated sync:tools when prod or dev tools/list changes.
Generated Friendly Commands
Every checked-in generated catalog tool is reachable through its deterministic command path. Scalar schema fields are exposed as flags, with both kebab-case and schema-case accepted:
ainecto task list-tasks --env dev --document-uuid <documentUuid>
ainecto task list-tasks --env dev --documentUuid <documentUuid>
Array or object payloads use the existing first-party CLI JSON payload reader:
ainecto documents create --env dev -f create-documents.json
cat task-ops.json | ainecto task apply-changes --env dev --json
Destructive generated commands prompt in human mode unless --yes is supplied. In --json mode they fail with a structured error unless --yes is present.
Attachment file upload is available through a bespoke command that performs the upload-token, raw PUT, and attachment registration flow:
ainecto attachments upload --env dev --document-uuid <documentUuid> ./diagram.png ./notes.pdf
The generated request_upload_token and upload_attachments paths remain raw MCP argument-contract commands and do not read local files.
Local Tarball Smoke
npm pack
npx -y ./ainecto-mcp-0.1.4.tgz --help
npm exec --package ./ainecto-mcp-0.1.4.tgz -- ainecto --help
Live Dev MCP Smoke
This requires an authenticated dev token from ainecto auth login --env dev or a local developer AINECTO_TOKEN.
npm run smoke:mcp -- --env dev
The smoke initializes MCP, reads tools/list, calls mcp__ainecto__list_projects, and prints only aggregate metadata such as tool count and Task tool exposure.
Local Attachment Upload Smoke
This exercises the full attachments upload flow against a local ainecto-api dev server:
OAuth MCP token issue, REST fixture document creation, CLI upload, list_attachments, and server filesystem byte verification.
npm run smoke:attachments -- --base-url http://localhost:8080 --endpoint http://localhost:8080/mcp --api-root /Users/ryan/project/workspace/codelive/ainecto-api
The smoke creates a temporary user/workspace/project/document and removes the fixture account, local temp file, and stored upload file unless --keep is supplied.
Known Limitations
- The MCP connector currently implements newline-delimited stdio JSON-RPC to HTTP JSON-RPC. Streamable HTTP SSE responses and
MCP-Session-Idsession handling are not implemented in Phase 1.
Official MCP Registry publish is pending GitHub namespace owner authentication with mcp-publisher and metadata submission.
推荐服务器
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。
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。
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。
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。
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。
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。
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。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
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。