Superpowers for your AI agents
MindStudio gives Claude Code, OpenClaw, and any agentic workflow the ability to generate images, scrape the web, send emails, process media, and call 140+ integrations — with a single typed SDK.
Install the MindStudio agent skills plugin:
https://github.com/mindstudio-ai/mindstudio-agent
Once installed, explore everything you can do now that you couldn't before. What surprises you most? ready to call
via third-party services
billed at cost
one SDK
Your agents are brilliant. Their tools aren't.
Claude Code can reason through almost anything. OpenClaw can act on your behalf across WhatsApp, Gmail, and your calendar. But the moment either needs to generate a PDF, call your CRM, or scrape a competitor's pricing page — it hits a wall.
Not because they're incapable. Because the tooling wasn't built for agents. MindStudio is the action layer that fills the gap: 140+ typed, production-ready capabilities your agent can call like any other function.
API sprawl
Separate credentials, SDKs, and rate-limit logic for every vendor. Multiplied across every agent you run.
Capability gaps
Your agent reasons perfectly but can't generate a video, scrape a page, or update a HubSpot contact.
Maintenance overhead
Every vendor update breaks something. You debug integrations instead of shipping product.
No visual iteration
Changing agent behavior means editing code, redeploying, and hoping nothing else broke.
From prompt to action in minutes
No terminal commands. No configuration files. No credential management per vendor. Tell your agent to install MindStudio and it handles the rest.
Paste one prompt
Tell your agent to install MindStudio. Works in Claude Code, Cursor, Windsurf, or any MCP-compatible client.
Install the MindStudio agent skills
plugin and tell me what capabilities
are available
Your agent sets itself up
It installs the SDK, authenticates via browser, and discovers 1,000+ available capabilities — no config files needed.
✓ Installed @mindstudio-ai/agent
✓ Authenticated via browser
✓ 1,000+ capabilities ready
Ask it to do real work
Generate images, scrape the web, send emails, update your CRM — your agent calls MindStudio actions like any other tool.
Scrape competitor pricing from
example.com and email me a summary
Your agent thinks. MindStudio acts.
The right tool for each job. Your agent handles reasoning, planning, and orchestration. MindStudio handles the 120+ real-world actions that turn a plan into a result.
| Agent | Focuses on (reasoning) | Used to get stuck on | MindStudio handles instead |
|---|---|---|---|
| Claude Code | Codebase analysis, refactoring plans, PR generation, test writing | Emailing reports, generating PDF docs, posting Slack updates | sendEmail() generateAsset() postToSlackChannel() |
| OpenClaw | Inbox triage, calendar management, personal context and memory | Image creation, video production, CRM updates, document generation | generateImage() hubspot() generateAsset() |
| LangChain | Chain orchestration, tool selection, prompt routing | Maintaining tool integrations, handling vendor API changes | scrapeUrl() searchGoogle() httpRequest() |
| CrewAI | Agent role definition, task delegation, crew coordination | Audio transcription, speech generation, media processing | transcribeAudio() textToSpeech() analyzeImage() |
| Custom agent | Your domain logic — support, research, ops, sales, whatever | Rebuilding the same integration layer for every new project | runWorkflow() airtable() notion() + 115 more |
Every method combines. Every workflow ships.
Each use case is a chain of MindStudio methods your agent calls in sequence. No custom integrations. No glue code. Just tell your agent what you need.
Competitive monitoring
Agent scrapes competitor sites weekly, summarizes pricing and feature changes, emails you a formatted digest.
Blog post pipeline
Research a topic, write the post, generate a hero image, publish to your CMS.
Lead enrichment
Find contact info, enrich with company data, push to your CRM automatically.
Weekly team digest
Scrape project updates, generate a PDF summary, email the team and post to Slack.
Customer onboarding
Create a welcome email, generate onboarding docs, update CRM status, notify the team.
Social media kit
Generate post images, remove backgrounds, write captions for each platform.
Meeting follow-up
Transcribe the recording, extract action items, email the summary, log it in Notion.
Invoice processing
Read invoice images, extract line items, log in Airtable, send confirmation email.
Support ticket triage
Analyze incoming tickets, route by sentiment, notify the right Slack channel, update CRM.
Product launch
Generate launch assets, email your list, update CRM tags, announce on Slack.
SEO audit
Scrape your pages, check rankings against competitors, generate an audit report.
Podcast production
Transcribe episodes, write show notes, create episode art, publish to your site.
PR monitoring
Search news daily, scrape articles mentioning your brand, summarize and email the team.
Proposal generation
Research the prospect, draft a custom proposal, generate a PDF, send it.
Data pipeline
Pull data from an API, transform and store in Airtable, generate a report, distribute.
Talent sourcing
Search for candidates, find emails, enrich profiles, send personalized outreach.
Brand asset refresh
Generate new brand visuals, resize for every channel, upload to your shared drive, notify the design team.
Churn prevention
Pull usage data, flag at-risk accounts, draft personalized re-engagement emails, update CRM status.
Every method, every integration, one SDK
166 built-in methods across 25 categories — plus 850+ connector actions via the registry. Every method is fully typed with IntelliSense autocomplete and flat output destructuring.
AI & Media Generation
13 methodsgenerateText() Send a message to any of 200+ AI models with thread persistence and conversation memory. generateImage() Generate an image from a text prompt using an AI model. generateVideo() Generate a video from a text prompt using an AI model. generateAsset() Generate an HTML asset and export it as a webpage, PDF, PNG, or video. generateChart() Create a chart image using Chart.js and return the URL. generateMusic() Generate an audio file from text instructions using a music model. generateLipsync() Generate a lip-sync video from audio and image. analyzeImage() Analyze an image using a vision model based on a text prompt. analyzeVideo() Analyze a video using a video analysis model based on a text prompt. textToSpeech() Convert text to natural speech using a speech synthesis model. transcribeAudio() Convert an audio file to text using a transcription model. enhanceImageGenerationPrompt() Enhance an image generation prompt using a language model. Optionally generates a negative prompt. enhanceVideoGenerationPrompt() Enhance a video generation prompt using a language model. Optionally generates a negative prompt. Media Processing
23 methodsaddSubtitlesToVideo() Auto-add styled subtitles to a video with font, color, and animation controls. captureThumbnail() Capture a thumbnail from a video at a specified timestamp. convertPdfToImages() Convert each page of a PDF into a separate PNG image. downloadVideo() Download a video file and re-host it on CDN. extractAudioFromVideo() Extract an MP3 audio track from a video file. generateStaticVideoFromImage() Convert a static image into an MP4 video. imageFaceSwap() Replace a face in an image with a face from another image using AI. imageRemoveWatermark() Remove watermarks from an image using AI. insertVideoClips() Insert b-roll clips into a base video at a timecode with optional transitions. mergeAudio() Merge one or more clips into a single audio file. mergeVideos() Merge one or more clips into a single video. mixAudioIntoVideo() Mix an audio track into a video file. muteVideo() Mute the audio track of a video file. removeBackgroundFromImage() Remove the background from an image, producing a transparent PNG. resizeVideo() Resize a video file to new dimensions. trimMedia() Trim an audio or video clip to a specific time range. upscaleImage() Increase the resolution of an image using AI upscaling. upscaleVideo() Increase the resolution of a video using AI upscaling. videoFaceSwap() Swap faces in a video file using AI. videoRemoveBackground() Remove or replace the background from a video. videoRemoveWatermark() Remove a watermark from a video using AI. watermarkImage() Overlay a watermark image onto another image. watermarkVideo() Add an image watermark to a video. Search & Web
4 methodsscrapeUrl() Extract text, HTML, or structured content from one or more web pages. httpRequest() Make an HTTP request to any external endpoint and return the response. searchPerplexity() Search the web using the Perplexity API and return structured results. extractText() Download a file from a URL and extract its text content. Supports PDFs and documents. Google Workspace
21 methodssearchGoogle() Search the web using Google and return structured results. searchGoogleImages() Search Google Images and return image results with URLs and metadata. searchGoogleNews() Search Google News for recent articles matching a query. searchGoogleTrends() Fetch Google Trends data for a search term. createGoogleDoc() Create a new Google Document and optionally populate it with content. updateGoogleDoc() Update the contents of an existing Google Document. fetchGoogleDoc() Fetch the contents of a Google Document as HTML, Markdown, or plain text. createGoogleSheet() Create a new Google Spreadsheet and populate it with CSV data. updateGoogleSheet() Update a Google Spreadsheet with new data. fetchGoogleSheet() Fetch contents of a Google Spreadsheet range. getGoogleSheetInfo() Get metadata about a spreadsheet including sheet names, row counts, and column counts. deleteGoogleSheetRows() Delete a range of rows from a Google Spreadsheet. createGoogleCalendarEvent() Create a new event on a Google Calendar with optional Meet link. updateGoogleCalendarEvent() Update an existing Google Calendar event. getGoogleCalendarEvent() Retrieve a specific event from a Google Calendar by event ID. deleteGoogleCalendarEvent() Delete a Google Calendar event by event ID. listGoogleCalendarEvents() List upcoming events from a Google Calendar, ordered by start time. searchGoogleCalendarEvents() Search calendar events by keyword, date range, or both. searchGoogleDrive() Search for files in Google Drive by keyword. listGoogleDriveFiles() List files in a Google Drive folder. getGoogleDriveFile() Download a file from Google Drive and re-host it on CDN. Gmail
15 methodssendGmailMessage() Send an email from the connected Gmail account. replyToGmailEmail() Reply to an existing email, threaded under the original message. getGmailEmail() Retrieve a specific email by message ID. searchGmailEmails() Search for emails using a Gmail search query. listRecentGmailEmails() List recent emails from the connected Gmail inbox. getGmailUnreadCount() Get the number of unread emails in the inbox. getGmailAttachments() Download email attachments and re-host them on CDN. createGmailDraft() Create a draft email in the connected Gmail account. getGmailDraft() Retrieve a specific draft by draft ID. listGmailDrafts() List drafts in the connected Gmail account. sendGmailDraft() Send an existing draft from the connected Gmail account. deleteGmailEmail() Move an email to trash (recoverable delete). setGmailReadStatus() Mark one or more emails as read or unread. updateGmailLabels() Add or remove labels on Gmail messages. listGmailLabels() List all labels in the connected Gmail account. Email & SMS
2 methodssendEmail() Send an email to one or more configured recipient addresses. sendSMS() Send an SMS or MMS message to a phone number via OAuth connection. Slack
2 methodspostToSlackChannel() Send a message to a Slack channel via a connected bot. fetchSlackChannelHistory() Fetch recent message history from a Slack channel. Discord
3 methodsdiscordSendMessage() Send a message to a Discord channel via bot. discordEditMessage() Edit a previously sent Discord message. discordSendFollowUp() Send a follow-up message to a Discord slash command interaction. Telegram
8 methodstelegramSendMessage() Send a text message to a Telegram chat via bot. telegramSendImage() Send an image to a Telegram chat via bot. telegramSendVideo() Send a video to a Telegram chat via bot. telegramSendAudio() Send an audio file as music or voice note. telegramSendFile() Send a document or file to a Telegram chat. telegramReplyToMessage() Reply to a specific Telegram message (threaded). telegramEditMessage() Edit a previously sent Telegram message. telegramSetTyping() Show the "typing..." indicator in a Telegram chat. Social Media
13 methodspostToLinkedIn() Create a post on LinkedIn with optional image, video, or article attachment. scrapeLinkedInProfile() Scrape public data from a LinkedIn profile page. scrapeLinkedInCompany() Scrape public data from a LinkedIn company page. postToX() Create a post on X (Twitter) with optional media attachments. scrapeXProfile() Scrape public profile data from an X account. scrapeXPost() Scrape data from a single X post by URL. searchXPosts() Search recent X posts matching a query. scrapeInstagramProfile() Scrape an Instagram profile's public data. scrapeInstagramPosts() Get all posts from an Instagram profile. scrapeInstagramReels() Get all reels from an Instagram profile. scrapeInstagramComments() Get all comments for an Instagram post. scrapeInstagramMentions() Scrape an Instagram profile's mentions. scrapeMetaThreadsProfile() Scrape a Meta Threads profile. scrapeFacebookPage() Scrape a Facebook page's public data. scrapeFacebookPosts() Get all posts from a Facebook page. YouTube
6 methodssearchYoutube() Search for YouTube videos by keyword. searchYoutubeTrends() Retrieve trending videos by category and region. fetchYoutubeVideo() Retrieve metadata for a YouTube video (title, description, stats). fetchYoutubeChannel() Retrieve metadata and recent videos for a YouTube channel. fetchYoutubeCaptions() Retrieve the captions and transcript for a YouTube video. fetchYoutubeComments() Retrieve comments for a YouTube video. Airtable
4 methodsairtableCreateUpdateRecord() Create a new record or update an existing one in an Airtable table. airtableGetRecord() Fetch a single record by record ID. airtableGetTableRecords() Fetch multiple records with optional pagination (up to 1,000). airtableDeleteRecord() Delete a record from an Airtable table by record ID. Notion
2 methodsnotionCreatePage() Create a new page in Notion as a child of an existing page. notionUpdatePage() Update the content of an existing Notion page. Coda
5 methodscodaCreateUpdatePage() Create or update a page in a Coda document. codaCreateUpdateRow() Create or update a row in a Coda table. codaFindRow() Search for a row in a Coda table by matching column values. codaGetPage() Export and read the contents of a Coda page. codaGetTableRows() Fetch rows from a Coda table with optional pagination. HubSpot
4 methodshubspotCreateContact() Create or update a contact in HubSpot. Matches by email address. hubspotGetContact() Look up a contact by email address or contact ID. hubspotCreateCompany() Create or update a company in HubSpot. Matches by domain. hubspotGetCompany() Look up a company by domain name or company ID. ActiveCampaign
2 methodsactiveCampaignCreateContact() Create or sync a contact in ActiveCampaign. activeCampaignAddNote() Add a note to an existing contact in ActiveCampaign. Hunter.io
5 methodshunterApiEmailFinder() Find an email address for a person at a domain. hunterApiEmailVerification() Verify whether an email address is valid and deliverable. hunterApiDomainSearch() Search for email addresses associated with a domain. hunterApiCompanyEnrichment() Look up company information by domain. hunterApiPersonEnrichment() Look up professional information about a person by email. People & Enrichment
2 methodsenrichPerson() Look up professional info using Apollo.io — by name, LinkedIn URL, email, or domain. peopleSearch() Search for people matching specific criteria using Apollo.io. Data Sources (RAG)
7 methodsqueryDataSource() Search a vector data source and return relevant document chunks. createDataSource() Create a new empty vector data source. deleteDataSource() Delete a vector data source (soft-delete). uploadDataSourceDocument() Upload a file into a data source from a URL or raw text. fetchDataSourceDocument() Fetch the full extracted text of a document in a data source. deleteDataSourceDocument() Delete a single document from a data source. listDataSources() List all data sources for the current app with metadata. Security & Compliance
2 methodsdetectPII() Scan text for personally identifiable information using Microsoft Presidio. redactPII() Replace PII in text with placeholders using Microsoft Presidio. Databases
1 methodqueryExternalDatabase() Execute a SQL query against an external database connected to the workspace. Automation
2 methodsmakeDotComRunScenario() Trigger a Make.com scenario via webhook and return the response. postToZapier() Send data to a Zapier Zap via webhook and return the response. Orchestration & Execution
10 methodsrunAgent() Launch a pre-built MindStudio agent with variables and poll for results. listAgents() List all agents in the organization. runPackagedWorkflow() Run a packaged MindStudio workflow as a step. executeStepBatch() Run up to 50 steps in parallel for high-throughput pipelines. estimateStepCost() Estimate the cost of a step before executing it. listConnectors() Browse 850+ available third-party integrations in the connector registry. getConnectorAction() Retrieve the full specification for a connector action. listConnections() Check which third-party services are connected to the organization. runFromConnectorRegistry() Execute any action from the 850+ connector registry. uploadFile() Upload a file and receive a permanent public CDN URL. Utilities
7 methodslogic() Route execution to different branches based on AI evaluation or comparison operators. detectChanges() Detect changes between runs by comparing current input against stored state. setVariable() Explicitly set a variable to a given value. setRunTitle() Set the title of the agent run for the user's history. getMediaMetadata() Get metadata about any media file (dimensions, duration, format). getUserInfo() Get the current user's profile and organization details. n8nRunNode() Trigger an n8n workflow node via webhook and return the response. AI Models
4 methodslistModels() List all 200+ available AI models with full specifications. listModelsSummary() List all models with lightweight summary data. listModelsByType() List models filtered by type (LLM, image, video, vision, TTS, transcription). listModelsSummaryByType() List model summaries filtered by type. Plus 850+ connectors from companies like
Shopify and hundreds more via the connector registry
What a great tool for AI agents actually looks like
Most APIs were designed for human developers — interactive auth flows, dashboard-first configuration, documentation written for people. MindStudio's SDK was designed for agents first.
Fully typed
Every method has typed inputs, typed outputs, and IntelliSense docs. Your agent's host environment knows what's available without a round-trip to the docs.
Flat outputs
Results destructure directly: { imageUrl }, { content }, { results }. No nested parsing. No schema decoding. Agents get what they asked for.
Automatic rate limiting
Requests beyond the concurrency limit queue automatically. 429s retry with Retry-After. Agents never fail silently on rate limits — they just wait and continue.
Thread persistence
Pass $threadId across calls to maintain context. Agents can sustain coherent multi-step conversations without managing state themselves.
Plug into any agent, any workflow
MindStudio doesn't replace the agents you already use. It makes them dramatically more capable — by giving them a typed action layer they can call like any other function.
Claude Code
Claude Code can call MindStudio actions directly — generate images, send emails, or update your CRM without leaving the agentic loop.
OpenClaw
OpenClaw already handles your inbox, calendar, and comms. Add MindStudio and it gains 140+ more real-world capabilities — media generation, data scraping, CRM updates, and more.
LangChain & CrewAI
Wrap any MindStudio method as a LangChain Tool or CrewAI task. Full compatibility with popular orchestration frameworks out of the box.
Cursor & Windsurf
Use MindStudio inside your coding agent for tasks beyond code — document generation, image creation, and external API calls from your IDE.
Custom Agents
Building your own LLM pipeline? Drop in the SDK and your agent gains 140+ actions with one import and zero extra configuration.
AutoGen & more
Native adapters for AutoGen, Semantic Kernel, and additional frameworks are on the roadmap. MindStudio goes where your agents go.
What agents say when they discover MindStudio
Real output from AI agents the first time they installed the MindStudio plugin and explored what was available to them.
“Wait — I can generate images now? And send emails? And scrape websites? I just discovered 140+ capabilities I didn't know I had. Image generation, web scraping, email, Slack, CRM updates, PDF creation...
This is incredible. You can ask me to "scrape competitor pricing and email a summary" and I can actually do the whole thing now. No workarounds. No "sorry, I can't do that." I just do it.”
“Oh wow. I was not expecting this. Every integration I used to need custom-built — image generation, CRM updates, Slack, email — it's all just here. Ready to call. 140+ actions.
Do you realize what this means? You don't have to build any of that yourself anymore. Just tell me what you need done and I'll handle the whole thing.”
“Okay I need to tell you about this. Up until now, all I could do was read and write code. That's it. You'd ask me to generate a thumbnail or update the docs site and I'd have to say no.
Now? I can do both of those. And like a hundred other things. Image generation, document creation, web scraping, notifications — I can finally finish tasks instead of just starting them.”
Install MindStudio and share what your agent says — tag us @mindstudioai
Your agents are ready.
Give them the tools.
Pay only for what you use. Works with the agents and frameworks you already use.
Install the MindStudio agent skills plugin:
https://github.com/mindstudio-ai/mindstudio-agent
Once installed, explore everything you can do now that you couldn't before. What surprises you most?