Botster
Botster automates web data collection through configurable bots, job lifecycle management, and structured run results.
Verdict
Common use cases
- Scrape competitor pricing from e-commerce sites
- Extract lead contact details from LinkedIn searches
- Monitor Google Maps reviews for local businesses
- Pull Instagram engagement metrics for influencer campaigns
- Harvest product data from marketplaces at scale
Integration
- Vendor
- Botster
- Category
- other
- Auth
- API_KEY
- Tools
- 11
- Composio slug
botster
Tools
- Calculate Bot Cost
Calculate the credit price of one intended bot launch using the same input and bot-specific options, without launching the bot or spending credits.
- Delete Jobdestructive
Permanently delete one Botster job. This destructive action cannot be undone, is not the same as archiving, and may remove access to the job's run history.
- Delete Rundestructive
Permanently delete one Botster data run and its stored results without deleting the entire job. This destructive operation cannot be undone.
- Get Credit Balance
Return the connected Botster account's available credit balance; use this before launching or restarting credit-consuming work.
- Get Job
Get one job's status and data-run identifiers required to retrieve run results. The complete provider job object is returned without discarding bot-specific fields.
- Get Run Results
Return the complete bot-specific JSON for one data run identified from GET_JOB in the results field. The provider's result shape is preserved.
- Launch Bot
Launch a Botster bot, creating external work that may spend account credits. Use CALCULATE_BOT_COST with the same input and options first when cost matters. Returns launch metadata and normalizes an identifiable job identifier to job_id.
- List Bots
List or search bots available to the connected account and return their identifiers for cost calculation and launch.
- List Jobs
List jobs belonging to the connected Botster account and return job identifiers for inspection or lifecycle actions.
- Restart Job
Restart an existing Botster job, creating renewed external work that may repeat the bot's effects and spend additional account credits. The renewed job identity may differ from the original; returns all provider restart metadata.
- Set Job Archived
Archive a Botster job or restore it from the archive by setting its desired archived state. This reversible action changes organization state but does not delete the job.
Setup
Setup guide
- 11. Open your Switchy workspace and navigate to Settings → Integrations → MCP Servers. 2. Click 'Add MCP Server' and select Botster from the catalog. 3. Log in to your Botster account at botster.io, go to Settings → API, and generate a new API key. 4. Paste the API key into the Switchy connection form and click 'Connect'. 5. Switchy will verify the key and confirm the connection. 6. Open any Project, type '@Botster list available bots' in the chat, and press send. 7. If the MCP returns a list of bots with IDs, the connection works. 8. To launch a bot, first calculate cost with '@Botster calculate the cost of running [bot name] with [input parameters]', review the credit estimate, then launch with '@Botster launch [bot name] with [input]'.
What teammates see: by default, memories from Botster are scoped to the Project (PROJECT visibility) - you can mark any memory PRIVATE or share it ORG-wide.
Works well with
Top models
Compatibility data appears once enough Projects have used this MCP together with a given model.
How Switchy teams use it
Starter prompts
Check Credit Balance
@Botster what is my current credit balance? I need to know before I launch any new bots.
Find LinkedIn Scraper
@Botster list all available bots related to LinkedIn. I want to scrape profiles from a search query.
Estimate Scrape Cost
@Botster calculate the cost of running the Google Maps scraper for 500 business listings in San Francisco with full contact details.
Launch Instagram Bot
@Botster launch the Instagram hashtag scraper for #sustainablefashion, limited to the last 200 posts from the past 7 days.
Retrieve Scrape Results
@Botster get the results for job ID 12345. I need the full dataset as JSON so I can load it into our CRM.
Example outputs
Illustrative - representative of the model's voice and quality, not literal recordings.
@Botster list all available bots in my account and show me which ones can scrape LinkedIn profiles
Found 23 bots in your Botster account. Three are configured for LinkedIn: 'LinkedIn Profile Scraper' (extracts name, headline, experience), 'LinkedIn Company Scraper' (pulls company pages and employee counts), and 'LinkedIn Post Engagement' (tracks likes and comments on posts). Each bot shows its credit cost per run and last-used date. The Profile Scraper is your most-used bot with 47 historical runs.
This example shows LIST_BOTS filtering by keyword. The MCP returns bot identifiers you'll need for cost calculation and launch. Note that Botster's bot library changes as they add integrations—your account may have more or fewer bots depending on your plan tier and custom configurations.
@Botster calculate the credit cost to scrape 500 LinkedIn profiles with the Profile Scraper bot, then launch it if it's under 100 credits
Cost calculation complete: scraping 500 LinkedIn profiles will consume 85 credits at the current rate (0.17 credits per profile). Your account balance is 1,240 credits. Since 85 is under your 100-credit threshold, I've launched the bot. Job ID: job_8k3m9x. The scraper will run asynchronously and typically completes 500 profiles in 15–20 minutes. You can check progress with GET_JOB or wait for the run to finish.
This two-step workflow (CALCULATE_BOT_COST, then LAUNCH_BOT) prevents surprise credit burns. Botster bots run externally and consume account credits—always check cost first for large jobs. The MCP returns a job ID but does not poll for completion; you'll need a follow-up prompt to retrieve results once the run finishes.
@Botster get the results from my most recent Instagram hashtag scrape job and summarize the top 10 posts by engagement rate
Retrieved 340 posts from job 'Instagram Hashtag Scrape #craftbeer' (run completed 8 minutes ago). Top 10 by engagement rate: Post by @hoppybrews (4.2K likes, 890 comments, 12.3% engagement), Post by @brewmaster_daily (3.8K likes, 760 comments, 11.1% engagement)… The highest-performing posts feature carousel images of beer pours and behind-the-scenes brewery shots. Average engagement across all 340 posts is 6.7%, so these top 10 are performing 65–84% above baseline.
This example chains GET_JOB (to find the latest run ID) with GET_RUN_RESULTS, then asks the AI to analyze the raw JSON. Botster returns bot-specific result shapes—Instagram scrapes look different from LinkedIn scrapes. The MCP preserves the provider's schema, so the AI must parse varying structures depending on which bot ran.
Use-case deep-dives
When Botster beats manual scraping for cold outreach prep
A 3-person sales team running 200 cold emails per week needs LinkedIn profiles scraped and enriched before each batch. Botster's LIST_BOTS and LAUNCH_BOT tools let you queue scraping jobs from Switchy without logging into another dashboard. The CALCULATE_BOT_COST tool shows credit burn before you commit, which matters when you're paying per profile. The catch: if your team already owns Apify or a dedicated scraper, Botster adds a second vendor relationship for the same outcome. This MCP wins when you need pre-built scrapers for LinkedIn, Instagram, or Google Maps and don't want to maintain custom scripts. Check GET_CREDIT_BALANCE before each batch to avoid mid-run failures. If you're scraping under 50 profiles per week, the API overhead isn't worth it—just use the Botster web UI.
Botster for one-off competitive intelligence pulls
A product manager needs competitor pricing scraped from 12 e-commerce sites before a quarterly roadmap review. Botster's bot library covers common targets like Amazon, eBay, and Shopify storefronts without writing selectors. From Switchy, you call LIST_BOTS to find the right scraper, CALCULATE_BOT_COST to confirm budget, then LAUNCH_BOT and GET_RUN_RESULTS to pull the data into your workspace. The DELETE_RUN tool cleans up test runs so your job history stays readable. The limitation: Botster charges per run, so recurring weekly scrapes get expensive fast compared to a self-hosted solution. This MCP is the right call for ad-hoc research where setup time costs more than credits. If you're running the same scrape daily, build it in-house or use a flat-rate service.
When Botster's event scrapers save hours of manual copy-paste
A 2-person conference team needs attendee lists from Eventbrite and Meetup to plan follow-up campaigns. Botster's pre-built event scrapers pull names, emails, and company data in one shot. The GET_JOB and GET_RUN_RESULTS tools let you monitor progress and retrieve CSVs without switching tabs. The DELETE_JOB tool archives old events so your workspace doesn't bloat. The trade-off: Botster's event bots work best for public listings; if your target events require login or have anti-scraping measures, you'll hit rate limits or incomplete data. This MCP wins when you're scraping 3-10 events per quarter and need results in under an hour. If you're doing this monthly, negotiate a bulk credit package or evaluate a dedicated event-data API.
Frequently asked
What does the Botster MCP do in Switchy?
It lets your team launch Botster's web-scraping bots, check credit balances, and retrieve scraped data without leaving Switchy. You can calculate run costs before spending credits, list available bots, and manage jobs and runs. Think of it as a remote control for your Botster account inside your AI workspace.
Do I need admin access to connect Botster?
You need a Botster API key with permission to launch bots and read account data. Botster doesn't publish granular scope controls—the key either works or it doesn't. Whoever connects it should have billing visibility, because launching bots spends account credits and there's no undo.
Can the MCP scrape websites Botster doesn't support?
No. The MCP only launches bots Botster already built—LinkedIn, Instagram, Google Maps, and so on. If Botster's catalog doesn't include a site, the MCP can't scrape it. You'll need a different tool or a custom scraper. Use List Bots to see what's available before you plan a workflow.
Why use this instead of Botster's dashboard?
Speed and context. You can calculate costs, launch a bot, and pull results into a prompt without switching tabs or copy-pasting job IDs. The MCP also exposes the full provider JSON, so you can chain scraped data into other tools—Airtable, Notion, Slack—inside the same Switchy workflow.
Who on the team should connect Botster?
Whoever pays the Botster bill or manages scraping workflows. Launched bots spend real credits, and deleted jobs can't be recovered. If multiple people need access, connect one shared account and document which bots cost what. Switchy doesn't isolate credit usage per user.