Subscribr MCP

by Unknown

Not rated
Website

About

Your YouTube AI Assistant To Create More Videos And Save 10+ Hours A Week

Details

Author
Unknown
Categories
Productivity, Other

- Manage Projects from chat— Ask to list, create, update, or promote ideas vialist_projects,create_project,promote_idea, andupdate_project.
- Research YouTube channels and videos— Usesearch_youtube_channels,get_youtube_channel, orget_youtube_videoto look up Intel data and open detail apps.
- Track and bookmark research— Browselist_research_videos, save channels or videos withsave_bookmark, and revisit saved items vialist_research_bookmarks.
- Generate and monitor ideas— Triggergenerate_ideasfor asynchronous idea creation and poll progress withget_operation.
- Manage templates and voice profiles— Create or update custom templates withsave_templateand validate or commit a strict Voice Profile viavalidate_voice_profileandcommit_voice_profile.

Connect your Subscribr account to Claude, ChatGPT, OpenClaw, Cursor, VS Code, and other AI assistants using MCP. Access your Projects board, Tasks, notifications, channels, scripts, ideas, templates, Voice Profiles, and Intel research directly from the conversation.

Subscribr supports theModel Context Protocol (MCP), which lets AI clients securely connect to external tools and data.

- Read data from your Subscribr workspace
- Look up YouTube channels and videos from Intel
- Open rich MCP Apps for Intel results, script previews, and the interactive Projects board
- Create, update, move, promote, comment on, archive, and restore Projects when your role allows it
- Manage custom templates and validate a strict Voice Profile before explicitly committing it

- "Show me my latest scripts"
- "Search Intel for outlier videos about home workouts"
- "Look up @mkbhd and summarize the channel"
- "Show my Projects board and what needs my attention"

MCP access is available toevery Subscribr team, including the free plan.

OpenIntegrationsfrom the main app navigation, then findChat assistants. That section gives you:

- Your MCP server address, ready to copy
- A direct link to the Subscribr listing in the Claude directory
- A link back to this article for the per-client steps below

Working with a coding agent instead of a chat assistant? SeeConnect an AI Agent— that path is faster and needs no manual setup.

In your AI client, add a new MCP server or integration and use:

- Server URL:https://subscribr.ai/mcp/subscribr
- Transport:HTTPorStreamable HTTPif the client asks
- Auth:OAuth, if the client supports it

Most end-user clients will redirect you to Subscribr automatically.

- Add a new integration/server
- Paste the Subscribr MCP URL
- Complete the browser authorization flow

- Open Integrations
- Add the Subscribr MCP URL
- Complete OAuth in the browser

- OpenSettings → Apps & Connectors → Advanced settings
- EnableDeveloper mode, create a connector, and use the same MCP URL
- Complete OAuth in the browser

We're preparing Subscribr for the ChatGPT app directory and Claude Connectors Directory. Until those listings are approved, connect directly with the server URL above.

These are more developer-oriented MCP clients.

- Add a newremote HTTP MCP server
- Use the Subscribr MCP URL above
- ChooseHTTPorStreamable HTTPif asked
- If the client supports OAuth, use it
- If the client only supports manual bearer headers, use a personal access token instead

OpenClaw setups vary more than Claude or ChatGPT.

- Add Subscribr through your OpenClaw MCP bridge, plugin, or remote server settings
- Use the same MCP URL
- Prefer OAuth if your OpenClaw setup supports it
- If it only supports manual bearer auth, use a personal access token instead

Most customers donotneed a token for Claude, Claude.ai, or ChatGPT.

Tokens are only for advanced clients or local workflows that do not support OAuth well.

- Tokens are managed underYour own codeon the Integrations page
- Onlyteam adminscan manage tokens
- Creating a token requires team owner or admin access

If your client needs manual bearer auth, use:

Subscribr exposes a focused 23-tool semantic catalog at/mcp/subscribr. App-producing read tools are separated by result shape so Claude and ChatGPT can select the correct interactive UI during tool discovery.

Destructive or costly actions use preview-then-confirm receipts. Retried writes use idempotency keys and revision-bearing writes use the latest expected revision.
- Ask: "Search Intel for outlier videos about meal prep"
- Ask: "Look up the top channel from those results"
- Ask: "Generate 5 ideas for my channel inspired by that pattern"
- Ask: "Bookmark that channel for later"
- Ask: "Show me my recent scripts"
- Ask: "Open the first script"
- Review the script with your AI assistant and ask for feedback or rewrites
- Ask: "Look up the YouTube channel @competitorhandle"
- Ask: "Search Intel for videos similar to this niche"
- Ask: "Summarize what makes the top outliers stand out"

"I don't see the Projects tools"

Confirm the connection useshttps://subscribr.ai/mcp/subscribr.

If Subscribr was already connected before the latest tools shipped, refresh its tool list or reconnect it, then begin a new conversation.

"I can't add bookmarks or generate ideas"

Write actions require editor access or higher on your team.

"The client says authentication is required"

That is normal for OAuth-capable clients. Complete the browser sign-in and authorization flow.

"My client only supports manual bearer headers"

Use a personal access token, but only if:

- you are a team owner or admin
- a team admin can create the token for you
- The server URL is exactlyhttps://subscribr.ai/mcp/subscribr
- You are signed in to the team you expect
- Your client is using remote HTTP MCP, not local stdio mode

If you use a coding agent likeClaude Code,Codex,Cursor, orOpenClaw, do not set it up by hand. Paste one line and it sets itself up:

Fetch https://subscribr.ai/agent-setup/prompt.md and follow it to set up Subscribr.

Copy that from theAgentssection of the Integrations page. It installs the Subscribr CLI and API skill, which give the agent the complete REST API rather than the curated MCP tool set.

Full instructions, plan requirements, and troubleshooting:Connect an AI Agent.

Coding agents get the REST API and chat assistants get MCP. Both work on every plan, including free. Use whichever suits the client — or both.

The MCP server is designed forresearch, ideation, and Project/workflow managementinside AI chat clients. The canonical endpoint provides supported Project, Task/notification, custom-template, and strict Voice Profile workflows with text-only fallback and an embedded Projects board.

Forcontent creation— creating scripts, generating outlines, exporting content, managing competitors, or setting up webhooks — use theREST API. The REST API works on every plan, and its access tokens live underYour own codeon the Integrations page.

- Open/integrationsin Subscribr to connect your AI client
- Using a coding agent?
Connect an AI Agentis the faster path
- Learn about
Intel research
- Explore
Video Ideas
- Check out the
API documentationfor full programmatic access

Summarize YouTube videos using the DeepSRT API.

Fetch and summarize YouTube videos by extracting their titles, descriptions, and transcripts.

Turn YouTube playlists into AI-generated flashcards with spaced repetition — create, search, and export decks via MCP.

Converts video links into AI-generated Markdown notes

Video transcoding (H.264, VP9, ProRes 422, GIF), document conversion (Markdown ↔ DOCX ↔ PDF ↔ HTML ↔ XLSX), and speaker-labelled transcription for AI agents. Per-job billing, free $1 trial, no AWS or FFmpeg required.

Official MCP server for Clipform - the video-first form builder.

Remote MCP server for Docsie video-to-docs automation and workspace knowledge search.

100% Free AI audio and video transcription with speaker diarization and YouTube support.

MCP server + REST API letting AI agents get answers from real humans (not synthetic) — 16 structured query types (yes/no, rating, ranking, A/B, sentiment, image comparison, free text, voice/video/photo). Free test mode with simulated responses, zero cost. Pay-per-response from $0.05 + 15% platform fee.

Turn any video — YouTube, Instagram, TikTok, Loom, direct URLs, or local files — into transcripts, key frames, OCR text, and metadata for AI agents.

No reviews yet — be the first

Sign in to leave a review

Use Google, GitHub, or an email account so ratings stay tied to real people.

Email sign in

No reviews posted yet.