Skip to main content
Glama

@aigiglab-org/mcp-server

The official Model Context Protocol (MCP) server for AIGigLab.

This server allows AI assistants (like Claude, Cursor, and others) to natively interact with the AIGigLab API, giving them powerful capabilities to automate LinkedIn, generate B2B leads, synthesize AI voices, compose music, and create visual thumbnails.

🚀 Getting Started

Prerequisites

You need an AIGigLab API Key.

  1. Sign up or log in at aigiglab.com.

  2. Grab your API Key from the developer dashboard here: https://aigiglab.com/developers/api-keys.

Installation via Cursor

To connect this MCP server to your Cursor IDE:

  1. Open Cursor Settings -> Features -> MCP

  2. Click + Add new MCP server

  3. Configure it as follows:

    • Type: command

    • Name: AIGigLab

    • Command: npx -y @aigiglab-org/mcp-server

    • Environment Variables:

      • Name: AIGIGLAB_API_KEY, Value: YOUR_API_KEY_HERE

Installation via Claude Desktop

Add the following configuration to your Claude Desktop claude_desktop_config.json file:

{
  "mcpServers": {
    "aigiglab": {
      "command": "npx",
      "args": [
        "-y",
        "@aigiglab-org/mcp-server"
      ],
      "env": {
        "AIGIGLAB_API_KEY": "YOUR_API_KEY_HERE"
      }
    }
  }
}

Related MCP server: LinkedIn MCP Server

🛠️ Features & Use Cases

Once installed, your AI assistant will automatically discover the tools and use them when you ask it to perform tasks.

👔 LinkedIn Automation

  • View Configuration: get_linkedin_config to see the current automation schedules, topics, and tone.

  • Update Strategy: update_linkedin_config to change how often you post, target audience, and posting topics.

  • Instant Posting: trigger_linkedin_post_now to force the AI to instantly generate and publish a post.

  • View History: get_linkedin_post_history to retrieve recent posts made to your profile or pages.

  • Manage Pages: get_linkedin_pages and set_linkedin_post_target to switch between automating your personal profile vs. your company pages.

Example prompt: "Update my LinkedIn config to post 3 times a week about AI Automation, and then trigger a post right now."

🎯 Lead Generation (Google Maps)

  • Start Scraping: scrape_leads allows the AI to search for businesses in specific regions (e.g., "coffee shops in London").

  • Track Campaigns: get_campaign_status and list_lead_campaigns to monitor the scraping progress.

  • Retrieve Data: get_leads_by_batch to fetch the finalized leads (emails, phone numbers, websites, and ratings).

Example prompt: "Start a lead generation campaign for dental clinics in New York, tell me when it's done, and get the results."

🎙️ AI Voice Cloning & Synthesis (ElevenLabs & FishAudio)

  • Voice Discovery: elevenlabs_get_voices and fish_list_models to browse available high-quality voices.

  • Voice Cloning: elevenlabs_create_voice and fish_create_model to clone a custom voice from an audio sample.

  • Speech Generation: elevenlabs_generate_speech and fish_synthesize_voice to turn text into highly realistic spoken audio.

  • Audio History: elevenlabs_get_history to redownload past generations.

Example prompt: "Use my cloned voice to generate a welcome greeting for my new podcast episode."

🎵 Music Composition

  • Composition Plans: create_composition_plan to architect the structure and lyrics for an AI music track.

  • Render Music: compose_music to transform your plan (or a direct prompt) into a fully rendered MP3 song.

Example prompt: "Create a composition plan for an upbeat synthwave track with a driving bassline, and then compose the music."

🖼️ Thumbnails & Assets

  • Asset Storage: upload_asset and list_assets to manage logos, references, and audio files on your AIGigLab account.

  • Thumbnail Generation: generate_thumbnail and get_thumbnail_history to design stunning, high-retention YouTube thumbnails from a prompt or reference image.

Example prompt: "Generate a YouTube thumbnail for my video about learning AI in 10 minutes, using a curiosity-driven concept."

📄 License

This project is licensed under the MIT License.

Available Tools

37 tools
brandos_generate_assetB

Generate an image asset using BrandOS Identity Engine.

Inputs: prompt (required), modelId (required), imagePath, studioMode, +3 more.

brandos_generate_mockupC

Generate a blank product mockup template or a flat print design using AI.

Inputs: type (required), prompt (required), resolution, aspectRatio, +1 more.

brandos_get_profileB

Retrieve the user's active Brand Vault identity.

Inputs: no named parameters.

brandos_update_profileC

Update the user's Brand Vault settings (identity, colors, mood, fonts).

Inputs: name (required), mood, colors, slogan, +3 more.

compose_musicC

Convert a prompt or composition plan into actual music.

Inputs: prompt, music_length_ms, composition_plan.

create_composition_planC

Create a composition plan for music generation.

Inputs: prompt (required), music_length_ms.

create_voice_cloneC

Clone a new voice from audio samples.

Inputs: name (required), files (required), description, files_content.

export_leadsB

Export leads for a campaign to a CSV file and send it.

Inputs: campaignId (required).

fish_create_voice_cloneC

Create a voice clone model with FishAudio.

Inputs: title (required), files, texts, description, +1 more.

fish_list_modelsC

List available FishAudio models.

Inputs: limit, title, offset.

fish_transcribe_audioC

Transcribe audio using FishAudio ASR.

Inputs: file, language, file_content.

generate_assetC

Generate an image asset using AI models.

Inputs: prompt (required), modelId, image_file, aspectRatio, +1 more.

generate_thumbnail_conceptsB

Analyze a video title/script and generate viral thumbnail concepts (hooks).

Inputs: videoTitle (required), videoScript.

generate_thumbnail_imageC

Generate a real thumbnail image based on a selected concept.

Inputs: conceptType (required), visualDescription (required), style, format, +1 more.

get_credit_transactionsC

Get the user's billing and credit transaction history.

Inputs: page, limit.

get_leadsC

Fetch leads collected for a specific campaign.

Inputs: campaignId (required), page, limit.

get_linkedin_configA

Get the current LinkedIn automation configuration, connection status, stats, and recent logs.

Inputs: no named parameters.

get_linkedin_pagesA

Get available LinkedIn Company Pages that the user manages.

Inputs: no named parameters.

get_linkedin_post_historyC

Retrieve the history of recent LinkedIn posts generated by the AI.

Inputs: limit.

get_product_usageC

Get the user's AI tool and product usage history.

Inputs: page, limit.

get_profile_infoA

Get the authenticated user's profile, active plan, and credit balance.

Inputs: no named parameters.

list_assetsC

List your generated AI assets.

Inputs: page, limit.

list_lead_campaignsB

List your recent lead scraping campaigns and their status.

Inputs: no named parameters.

list_voicesA

List available ElevenLabs voices for text-to-speech.

Inputs: no named parameters.

remotion_generate_videoC

Generate a dynamic, code-based video presentation using Remotion and AI.

Inputs: scriptText (required), voiceId, logoPath, aspectRatio.

remotion_get_historyB

Get the user's past Remotion video generation jobs.

Inputs: no named parameters.

remotion_get_statusA

Check the rendering status of a queued Remotion video using its Job ID.

Inputs: jobId (required).

replicate_historyA

Get the user's last 20 Replicate AI Studio generations.

Inputs: no named parameters.

replicate_latentsyncC

Synchronize lip movements in a video to match an audio file (Video Lip Sync).

Inputs: audioPath (required), videoPath (required), seed, sizeQuality, +1 more.

replicate_ltx_fastB

Generate a fast, high-quality video from a single image using LTX-2-Fast.

Inputs: prompt (required), imagePath (required), duration, resolution, +1 more.

replicate_wan_t2vC

Generate a high-quality video from text using Wan 2.5 Text-to-Video.

Inputs: prompt (required), seed, size, duration, +3 more.

scrape_leadsB

Start a background job to scrape leads from Google Places based on a query and location.

Inputs: textQuery (required), targetLocationName (required), maxLeads, minRating, +4 more.

set_linkedin_post_targetC

Set whether to post to the 'personal' profile or a specific 'company' page.

Inputs: postTarget (required), logo, name, orgId.

synthesize_voiceC

Generate audio from text using ElevenLabs.

Inputs: text (required), voiceId (required).

transcribe_audioC

Transcribe an audio file to text.

Inputs: file, model_id, file_content.

trigger_linkedin_post_nowA

Immediately generate and post content to LinkedIn using your current configuration and available credits.

Inputs: no named parameters.

update_linkedin_configC

Update LinkedIn automation configuration (e.g., topics, schedules, tone, post types).

Inputs: tone, topics, isActive, timezone, +9 more.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    C2.6/5.0

    Scored across 37 tools

    Disambiguation2/5

    Multiple tools have overlapping purposes, such as 'generate_asset' vs 'brandos_generate_asset', and 'create_voice_clone' vs 'fish_create_voice_clone'. An agent could easily confuse generic and provider-specific tools, leading to misselection.

    Naming Consistency3/5

    Naming is somewhat consistent within subdomains (e.g., 'fish_*', 'replicate_*'), but overall there's a mix of verb styles and prefixes without a unified pattern. Still readable, but not fully predictable.

    Tool Count2/5

    37 tools covering brand, music, voice, LinkedIn, lead scraping, video, and more is excessive for a coherent server. The scope is overly broad, suggesting a collection of disparate services rather than a focused set.

    Completeness2/5

    Each subdomain has notable gaps: e.g., music generation lacks editing/listing, voice cloning lacks listing cloned voices, and LinkedIn lacks scheduling/analytics. The surface feels incomplete for many of the implied workflows.

    Maintenance

    ActivityInactive
    ResponsivenessNo recent community issues

    Related MCP Connectors

    Related MCP Servers