MCP servers for a content creation agent
The MCP server bundle to build an AI that produces images, voiceover, and researched copy.
3 servers · ~40 tools · Last updated June 17, 2026
TL;DR: A content agent creates assets and grounds them in real material. This bundle pairs image and audio generation with a scraper for research, so the agent can go from brief to draft visuals and voiceover.
Bottom line: start with Fal.ai MCP Server and add the rest as your needs grow. All 3 install together via the merged config below (~40 tools total).
Tool budget: this stack exposes about 40 tools. That's within Cursor's practical ~40-tool ceiling, so all servers can stay enabled together. Check your own config →
What's in the stack
One-click config
All 3 servers merged into a single block — pick your client and paste.
{
"mcpServers": {
"fal-ai-mcp-server": {
"command": "uvx",
"args": [
"--from",
"fal-mcp-server",
"fal-mcp"
],
"env": {
"FAL_KEY": "<your-fal-api-key>"
}
},
"elevenlabs-mcp-server-official": {
"command": "uvx",
"args": [
"elevenlabs-mcp"
],
"env": {
"ELEVENLABS_API_KEY": "<your-api-key>"
}
},
"firecrawl": {
"command": "npx",
"args": [
"-y",
"firecrawl-mcp"
],
"env": {
"FIRECRAWL_API_KEY": "<your-firecrawl-api-key>"
}
}
}
}Paste into ~/Library/Application Support/Claude/claude_desktop_config.json and fully restart Claude Desktop. Replace any placeholder keys/tokens with your own.
Capabilities this stack covers
FAQ
Can one agent do images and voice?
Yes — with a generation server per medium (e.g. fal for images, ElevenLabs for voice), the agent orchestrates both from a single brief.
How do I control cost?
Generation is metered per asset. Set provider-side limits and prefer cheaper models for drafts, higher-quality for finals.