mcpField
A
agent-media-toolkit

agent-media-toolkit

Remote
github.com/krakonjac300-pixelMedia & Design1Updated 15w ago
Repository

About agent-media-toolkit

Agent Media Toolkit is a hosted collection of media and data tools for AI agents, accessed through one endpoint. It covers web research and text tasks alongside audio and video processing, with some tools paid per call.

Key features

  1. 01Web search and news lookup
  2. 02Page extraction and summarization
  3. 03Translation and content moderation
  4. 04Script writing and text-to-voiceover
  5. 05Transcription and subtitle generation
  6. 06Video clip discovery and editing

Use cases

  1. 01Research a topic and summarize sources
  2. 02Translate or moderate text
  3. 03Create narration or subtitles from media
  4. 04Find and edit short clips from long videos

MCP support

Provides a remote MCP server over streamable HTTP at `/mcp/`. MCP clients connect to the endpoint to use the tools; some media tools are paid per call via x402.

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://podcast-shorts-a2a-production.up.railway.app/mcp/
Copy install command
Claude CodeTerminal
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
Claude Desktopclaude_desktop_config.json
Windsurf~/.codeium/windsurf/mcp_config.json

Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.

Tools

15
ParameterTypeRequiredDescription
querystringYes
max_resultsintegerNo

Details

From the project README

Agent Media Toolkit

15 media & data tools your AI agent can hire per call. One hosted endpoint — web search, news, web-page extraction, summarizing, translation, moderation, script writing, text-to-voiceover, transcription, subtitles, viral-clip discovery, and short-form video editing. No API key, no signup. Free over MCP for the lighter tools; premium media tools are pay-per-call in USDC on Base via x402.

This is a hosted MCP server — there is nothing to install or run. Point your MCP client at the endpoint above and the tools appear.

Add it to your MCP client

{
  "mcpServers": {
    "agent-media-toolkit": {
      "url": "https://podcast-shorts-a2a-production.up.railway.app/mcp/"
    }
  }
}

Works with any MCP client that supports remote streamable HTTP servers (Claude, Cursor, and MCP-compatible agent frameworks).

The tools

ToolTierWhat it does
searchfreeLive web search → title/url/snippet
newsfreeRecent news for a query, with dates + sources
extractfreeWeb page → clean main-content text
summarizefreeTL;DR any text or a URL (paragraph or bullets)
repurposefreeText → thread, summary, titles, hashtags
translatefreeTranslate text into any language
scriptfreeTopic → short-form video script (hook/beats/CTA)
moderatefreeContent-safety classification
media_infofreeMedia URL → metadata, no download
voiceoverfreeText → natural MP3 narration
package_postfreeClip → platform titles/description/hashtags
transcribe$0.25Audio/video → timestamped transcript
subtitles$0.05Audio/video → .srt / .vtt caption file
find_clips$1.50Long video → scored viral clip candidates
edit_clip$3.00Time range → finished 9:16 captioned short

Premium tools return a pointer to their paid x402 endpoint; an unpaid HTTP request answers 402 Payment Required with a USDC-on-Base price quote, the agent pays, and the result is returned.

Also available over plain HTTP (x402)

Every tool is also a pay-per-call HTTP endpoint. Example:

POST https://podcast-shorts-a2a-production.up.railway.app/search
{ "query": "agent commerce protocol", "max_results": 5 }
→ 402 Payment Required  ($0.02, USDC on Base)
→ pay + retry → JSON results

Example: connect from Python

See examples/client.py.

License

MIT — see LICENSE. The hosted service is operated by The Solo Stack; this repo documents how to use it.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.