Aryankeluskar Poke Video Mcp

ai.smitheryPublished by smithery.aiOfficial publisherv1.14.0Updated Oct 24, 2026

Search your Flashback video library with natural language to instantly find relevant moments. Get…

The publisher's own server: 222wcnm Bilistalkermcp

VerifiedStreamable HTTPWeb executableWeb Search & ScrapingMedia & Design

Overview

AI-generated overview

A remote MCP server that lets an assistant search your personal Flashback video library in natural language and return matching clips with descriptions and URLs

What you can do
Search a Flashback video library with natural-language queries and get ranked matching segments with AI-generated visual descriptions and audio transcripts, plus direct links to roughly 30-second clips and their relevance scores and URL expiry times. Retrieve setup and usage guidance for the video query system. Read information about the underlying video processing API as a resource.
When to use it
Useful when you already upload videos to Flashback and want an assistant to find specific moments in them by describing what happens, such as a presentation, a pet running, or a meeting topic, instead of scrubbing through footage manually.
Requirements
A remote endpoint reached over streamable HTTP with a bearer token in the Authorization header for Smithery authentication. A Flashback account user ID is needed for configuration, and access to the Flashback video processing system. Videos must already be uploaded to the account.
Before you install
Requires a bearer token in the Authorization header and your Flashback user ID, so treat both as sensitive. Queries and account identifiers are sent to the Flashback backend and third-party services including Pinecone, Anthropic Claude, OpenAI Whisper and Google Cloud Storage. Video links are presigned and expire after one hour.

Installation

In SourceWeft

  1. Open Aryankeluskar Poke Video Mcp in the dashboard and add it to a workspace.
  2. Connect Authorization token credentials. They are stored privately and never shown on public pages.
  3. Enable the server for the chats that should use its tools.

Web executable via Streamable HTTP. Remote servers run from the web runtime once configured in a workspace.

Use with an agent

Paste this prompt into Claude Code, Codex, Cursor or another agent — or into SourceWeft chat. The agent reads this server's install guide, shows you its publisher, endpoint and sign-in, and adds it to your client once you agree. Keys never go into a command.

Read https://sourceweft.com/mcp/ai.smithery%2Faryankeluskar-poke-video-mcp/install.md and connect the MCP server it describes to my client. Before changing any config, show me its publisher, endpoint and sign-in method and wait for my OK. Never put an API key in a command or in a file I would commit; ask me where it should come from.

Read the install guide the agent follows

Other MCP clients

Add this to your client's mcpServers config.

{
  "mcpServers": {
    "aryankeluskar-poke-video": {
      "type": "http",
      "url": "https://server.smithery.ai/@aryankeluskar/poke-video-mcp/mcp",
      "headers": {
        "Authorization": "Bearer <YOUR_TOKEN>"
      }
    }
  }
}

README

Video Query MCP Server

An MCP (Model Context Protocol) server that provides video search capabilities using natural language queries. This server interfaces with the Flashback video processing API to search through your personal video collection.

Features

  • 🔍 Natural Language Search: Query videos using everyday language (e.g., "person giving presentation", "dog running in park")
  • 🤖 AI-Generated Descriptions: Returns detailed descriptions of video content including visual analysis and audio transcription
  • 🎬 Video Clips: Get direct URLs to relevant 30-second video segments
  • ⚡ Fast & Secure: Presigned URLs with 1-hour expiration for secure access
  • 📊 Relevance Scoring: Results ranked by semantic similarity to your query

Setup

Prerequisites

  • Your Flashback account user ID
  • Access to the Flashback video processing system

Configuration

When connecting this MCP server to Poke or other MCP clients, you'll need to provide:

  • user_id: Your unique Flashback account identifier (e.g., 4087fce3-3d86-4047-b35f-4004b4c19192)

Development

  1. Run the server:

    bash
    uv run dev

  2. Test interactively:

    bash
    uv run playground

Usage Examples

Query videos:

query_videos("person talking", max_results=5)

Get setup help:

get_setup_instructions()

How It Works

  1. Video Processing: Videos uploaded to Flashback are automatically:

    • Split into ~30-second segments
    • Analyzed with AI for visual content
    • Transcribed for audio content
    • Stored in a searchable vector database

  2. Search Process: When you search:

    • Your query is converted to embeddings
    • The system finds matching video segments
    • Returns descriptions and URLs for relevant clips

  3. Results: Each result includes:

    • AI-generated description of the video content
    • Relevance score (0-1, higher = more relevant)
    • Direct URL to view the video segment
    • Expiration time for the URL

API Reference

Tools

query_videos(query: str, max_results: int = 10) -> str

Search for video clips based on natural language query.

Parameters:

  • query: Natural language description of what you're looking for
  • max_results: Maximum number of results to return (1-15)

Returns: Formatted text with video descriptions and URLs

get_setup_instructions() -> str

Get detailed setup instructions for the video query system.

Returns: Complete setup and usage guide

Resources

  • api://video-processing: Information about the underlying video processing API

Examples

python
# Search for specific contentquery_videos("meeting discussion about deadlines")query_videos("someone cooking in kitchen")query_videos("red car driving")
# Limit resultsquery_videos("presentation", max_results=3)

Troubleshooting

  • No results found: Check that videos have been uploaded to your Flashback account
  • "No description available": Older videos may need to be re-processed for full descriptions
  • Expired URLs: Video URLs expire after 1 hour for security - request fresh results if needed

Technical Details

  • Backend: FastAPI service deployed on Modal
  • Vector Database: Pinecone for semantic search
  • AI Models: Anthropic Claude for visual analysis, OpenAI Whisper for transcription
  • Storage: Google Cloud Storage for video files

Deploy

Ready to deploy? Push your code to GitHub and deploy to Smithery:

  1. Create a new repository at github.com/new

  2. Initialize git and push to GitHub:

    bash
    git add .git commit -m "Video Query MCP Server 🎬"git remote add origin https://github.com/YOUR_USERNAME/YOUR_REPO.gitgit push -u origin main

  3. Deploy your server to Smithery at smithery.ai/new

Source: README.md at commit 2152b0d

Tools

0
SourceWeft has not read this server's tools yet.

Version history

1
  1. v1.14.0LatestSep 16, 2026

Score

75% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 daysRequired25 pts
  • Auth declaredRequired10 pts
  • Tools listedRequired15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts

    Declare a license in the repository or the registry entry.

Other Smithery integrations

222wcnm Bilistalkermcp

ai.smitheryOfficial publisher

Bilibili creator analysis server that lets an assistant look up a specific user's profile, videos, dynamics, articles, subtitles, followings and comments.

Streamable HTTPWeb executable
Knowledge & MemoryNot indexed yetOct 24, 2026

Zwldarren Akshare One Mcp

ai.smitheryOfficial publisher

Remote MCP server that gives an assistant read-only access to Chinese stock market data: historical and real-time prices, news, and financial statements.

Streamable HTTPWeb executableAuth
FinanceCredential requiredOct 24, 2026

Zhaoganghao Hellomcp

ai.smitheryOfficial publisher

A small remote MCP server that lets an assistant greet people by name and share background on the origin of 'Hello, World'.

Streamable HTTPWeb executable
Not indexed yetOct 24, 2026

Zeta Chain Cli

ai.smitheryOfficial publisher

A remote MCP server that lets an assistant generate friendly, customizable greetings for any name or audience.

Streamable HTTPWeb executableAuth
Credential requiredOct 24, 2026

Yuna0x0 Hackmd Mcp

ai.smitheryOfficial publisher

An MCP server that connects an assistant to the HackMD API so it can read, create, update and delete HackMD notes and manage teams.

Streamable HTTPWeb executable
Communication & CollaborationNot indexed yetOct 25, 2026