ilo
A local-first MCP server for social publishing on X (Twitter), enabling draft management, scheduling, and posting via CLI or AI agents.
README
ilo
Agent-first X performance monitoring, drafting, scheduling, and publishing.
Give your agent the public posts and performance data you care about, find the patterns worth repeating, then draft and publish through ilo's CLI, local MCP server, agent skill, or TypeScript package.
Install
npm i -g iloso
ilo start
Create an X developer app with OAuth 2.0 read and write access, then register this callback:
http://127.0.0.1:8976/callback
ilo start opens the browser flow and stores OAuth tokens in your operating system keychain.
Use the CLI
ilo post --text "Hello from ilo"
ilo drafts create --text "Post this tomorrow"
ilo drafts list
ilo drafts schedule <draft-id> --at "tomorrow at 9am"
ilo scheduler watch
Publishing asks for confirmation. Scripts can pass --yes after getting approval elsewhere.
Add local MCP
claude mcp add ilo -- iloso mcp serve
codex mcp add ilo -- iloso mcp serve
Config-based clients can use:
{
"mcpServers": {
"ilo": {
"command": "iloso",
"args": ["mcp", "serve"]
}
}
}
The MCP server uses the same local database and keychain credentials as the CLI. Publishing tools require confirm: true.
Install the packaged agent skill with ilo skill install.
Use the TypeScript package
Install iloso in a Node 22 or newer project to call the same core functions directly:
npm install iloso
import { createDraft, scheduleDraft } from 'iloso'
const draft = createDraft('A post to review')
await scheduleDraft(draft.id, 'tomorrow at 9am')
Repository
packages/core: local config, keychain, SQLite, scheduling, and provider logicpackages/cli: theiloandilosocommandspackages/mcp: the local stdio MCP toolsapps/web: the statically generated ilo.so site and same-origin tool routes
Develop
pnpm install
pnpm lint
pnpm typecheck
pnpm test
pnpm build
The website can run locally with pnpm -C apps/web dev. See apps/web/README.md for its optional environment variables and Cloudflare deployment.
Roadmap
X publishing and scheduling are first. Bluesky and LinkedIn are the next provider targets. Local imports and automated performance reports come next.
License and support
MIT licensed. If ilo saves you time, sponsor Ian.
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。