# ai-resources [Health: Active]

**Category:** 💻 Developer Tools  
**Repository:** https://github.com/Drix10/ai-resources  
**GitHub Stars:** 122  
**Views:** 0  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/ai-resources

## Description
100,000+ AI briefings in 41 categories: models, dev tools, agents, research, funding. New briefings are fact-checked against their source post and link to it. Plain Markdown, CC BY 4.0. MCP server + Agent Skill let Claude, Cursor or VS Code ask for the latest. Read at blogs.drix10.com

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `npx` (confidence: high):

```json
"mcpServers": {
  "ai-resources": {
    "command": "npx",
    "args": ["-y","ai-resources-mcp"],
    "env": {
      "AI_RESOURCES_API_URL": ""
    }
  }
}
```

**Requires environment variables:** `AI_RESOURCES_API_URL` — the values above are empty placeholders; fill in real credentials before running (see the repository for what each one is for).

## Documentation

## What ai-resources does

ai-resources exposes a searchable archive of more than 100,000 AI briefings through the Model Context Protocol. The archive is organized into 41 categories, including AI models, developer tools, agents, research, funding, robotics, climate, quantum, security, and crypto.

Briefings are stored as plain Markdown and include dates, factual summaries, and links to the original post and any page referenced by that post. The collection is intended for engineers tracking technical changes, founders and investors following company activity, researchers and students monitoring publications, and developers assembling RAG, agent, or evaluation datasets.

## How it works

The package is published on npm as `ai-resources-mcp`. MCP clients start it as a local stdio process, normally through `npx`, which downloads the package when it is first run. The server reads from the AI Resources API, whose default base URL is `https://blogs.drix10.com`.

Each tool response includes source information and the attribution text for the archive, so an agent can preserve citations when answering questions. The underlying repository also offers the same material as raw Markdown, a JSON API, RSS, and `llms.txt`, although those access paths are separate from the MCP process.

## Setup and configuration

Node.js 18 or newer is required. The package can be launched without a manual installation:

```bash
npx -y ai-resources-mcp
```

For Claude Code, the documented command is:

```bash
claude mcp add ai-resources -- npx -y ai-resources-mcp
```

Claude Desktop, Cursor, and Windsurf use an MCP configuration with `npx` as the command and `-y ai-resources-mcp` as its arguments. VS Code uses a stdio server entry in `.vscode/mcp.json` or its user MCP configuration. On Windows systems where the client cannot invoke `npx` directly, configure `cmd /c npx -y ai-resources-mcp` instead.

Set `AI_RESOURCES_API_URL` to replace the default API base URL. Authentication is not required.

## Tools and capabilities

- `search_briefings(query, category?, limit?, since?)` performs keyword search and returns newest matches first. Every word in the query must match, and `since` accepts `YYYY-MM-DD`.
- `latest(category?, limit?)` returns recent briefings across the archive or within one category.
- `get_briefing(id)` retrieves the complete Markdown for a briefing identified by its URL or slug, including source links.
- `list_categories()` returns category slugs and their briefing counts.

These operations support questions such as what a company announced recently, which briefings belong to a category, or what funding events appeared during a specified period.

## Limitations and notes

Search is keyword-based rather than semantic: all supplied words must match. The archive is a collection of briefings, not an editor-selected shortlist or newsletter, so it serves a different purpose from curated recommendation lists. The repository content is licensed under CC BY 4.0, while the published MCP package is marked MIT.

_Full upstream README: https://allmcps.com/mcp/ai-resources/readme_