Agenty
Agenty is a web scraping and automation platform that enables users to extract data, monitor changes, and automate browser tasks without coding.
Verdict
Common use cases
- Clone a scraping agent for a new region
- Clear stale data before a fresh scrape run
- Add multiple agents to a project at once
- Disable an API key after a security audit
- Create a new list for incoming lead data
Integration
- Vendor
- Agenty
- Category
- other
- Auth
- API_KEY
- Tools
- 40
- Composio slug
agenty
Tools
- Add Agents to Project
Tool to add agent(s) to a project. use after confirming the project exists and you have valid agent ids. example: "add agents ['agent abc123','agent xyz789'] to project 123."
- Change API key status by ID
Tool to enable or disable an api key by its id. use when you need to toggle the activation state of an existing api key after retrieving its identifier.
- Clear List Rows
Tool to clear all rows in a list by its id. use when you need to wipe out all data in a list before re-populating (e.g., "clear all rows in list 123").
- Clone Agent by ID
Tool to clone an existing agent by its id. use when you need to duplicate an agent's configuration and optionally start the clone immediately.
- Create Agent
Tool to create a new agent. use when you need to add an agent with configuration and optional auto-start.
- Create API Key
Tool to create a new api key. use when you need to provision access credentials for the api.
- Create API Key
Tool to create a new api key. use when you need to provision access credentials for the api.
- Create List
Tool to create a new list. use after preparing the list name and optional description. example: "create a list named 'contacts' with description 'potential leads'."
- Create Project
Tool to create a new project. use when you need to initialize a project with a name and optional description or template.
- Delete Agent by IDdestructive
Tool to delete a single agent by its id. use when you need to permanently remove an agent after confirming it exists.
- Delete API key by IDdestructive
Tool to delete an api key by its key id. use when revoking access for a specific api key after confirming the correct key id.
- Delete List by IDdestructive
Tool to delete a specific list by its id. use when you need to permanently remove a list after confirming it exists.
- Download API keys
Tool to download all api keys under an account in csv format. use when you need a complete list of api keys for export or backup.
- Download job file by ID
Tool to download output files by job id. use when you need to fetch a specific output file after a job completes.
- Download Job Result by ID
Tool to download the agent output result by job id. use when you need to fetch the final output of a completed job in csv, tsv, or json format.
- Download jobs
Tool to download all jobs in csv format. use when you need a complete list of jobs for export or analysis.
- Download lists
Tool to download all lists in csv format. use when you need a complete set of lists for export or backup.
- Fetch all agents
Tool to fetch all active agents under an account. use after authenticating when you need to list agents with pagination and sorting.
- Fetch all jobs
Tool to fetch all jobs under an account. use when you need to view and filter the list of jobs with pagination and sorting.
- Fetch List Rows by ID
Tool to fetch all rows in a specified list. use when you need to retrieve all data entries from a list with optional pagination and sorting. example: "fetch rows from list '61db...' with limit=50".
- Get Agent by ID
Tool to fetch details of a specific agent by its id. use when you need full agent configuration before taking further actions.
- Get agent input by ID
Tool to get agent input by agent id. use when you need to inspect or validate an agent's current input configuration before updating or starting jobs.
- Get Agent Templates
Tool to fetch all public agent templates and sample agents. use when listing available templates for users to select.
- Get all API keys
Tool to retrieve all api keys under an account. use after authentication when you need to list api keys with pagination and sorting.
- Get all connections
Tool to get all connections. use after authenticating when you need to list connections with pagination and sorting.
- Get all lists
Tool to retrieve all lists under an account. use after authentication when you need to list lists with pagination and sorting.
- Get all projects
Tool to retrieve all projects under an account. use after authentication when you need to list projects with pagination and sorting.
- Get API key by ID
Tool to get an api key by key id. use when retrieving details of a specific api key after identifying its id.
- Get dashboard reports and usage
Tool to fetch account reports like pages used by agent, date, and product. use when analyzing usage over a date range.
- Get Job by ID
Tool to fetch details of a specific job by its id. use when you need the full job metadata before retrieving results.
- Get Job Logs by ID
Tool to fetch logs for a given job by its id. use when you need to review or debug job execution logs with pagination support.
- List job files by ID
Tool to list output files by job id. use after job completion when you need to discover available output files.
- Reset API key by ID
Tool to reset an api key by key id. use when you need to regenerate the secret for an api key after compromise or rotation.
- Start Agent Job
Tool to start a new agent job. use when you need to trigger execution of an existing agent.
- Stop Job by ID
Tool to stop a running job by job id. use when you need to halt an in-progress job before it completes. confirm the job id beforehand to avoid unintended stops.
- Update Agent by ID
Tool to update an agent's configuration and settings by agent id. use when you need to modify an existing agent after retrieving its current settings. example: "update agent 123 with new name and tags."
- Update API key by ID
Tool to update an api key by its id. use when you need to modify name, role, or enabled state after confirming the api key's identifier.
- Update Input by Agent ID
Tool to update agent input by agent id. use when you need to modify an agent's input source (url, manual data, list, or agent) after verifying the agent id.
- Update List by ID
Tool to update a list's name or description by list id. use when you need to modify list metadata after confirming its id. example: "update list 123 with name 'my list'".
- Upload CSV file to List
Tool to upload a csv file to a list. use when you need to bulk import rows into an existing list after verifying the list's column schema.
Setup
Setup guide
- 11. Open your Switchy workspace and navigate to Settings > Integrations. 2. Find Agenty in the MCP directory and click Connect. 3. Log in to your Agenty dashboard and go to Settings > API Keys. 4. Generate a new API key and copy it to your clipboard. 5. Paste the key into Switchy's connection dialog and click Authorize. 6. Switchy will verify the key and confirm the connection. 7. Open any Space and type '@Agenty list my agents' to test — you should see your existing scraping agents. 8. To invoke a tool, @mention Agenty and describe the task in plain language, like '@Agenty create a new agent named product-scraper' or '@Agenty clear all rows in list 456'. The MCP will map your request to the appropriate tool and execute it.
What teammates see: by default, memories from Agenty are scoped to the Space (PROJECT visibility) - you can mark any memory PRIVATE or share it ORG-wide.
Works well with
Top models
Compatibility data appears once enough Spaces have used this MCP together with a given model.
How Switchy teams use it
Starter prompts
Clone Production Agent
@Agenty clone agent abc123 and start it immediately so I can verify the new settings work.Open in a Space →
Clear List Before Import
@Agenty clear all rows in list 789 because we're about to run a new extraction and need a clean slate.Open in a Space →
Add Agents to Project
@Agenty add agents abc123 and xyz456 to project 101 so they're grouped under the same billing and monitoring.Open in a Space →
Create New Scraping Agent
@Agenty create a new agent named ecommerce-prices with auto-start enabled and default scraping settings.Open in a Space →
Disable Compromised Key
@Agenty disable API key key_9876 immediately because it was exposed in a public repository.Open in a Space →
Example outputs
Illustrative - representative of the model's voice and quality, not literal recordings.
@agenty create a new scraping agent called 'Product Monitor' that checks competitor pricing daily and auto-starts it
Agent 'Product Monitor' created successfully with ID agent_pm_8x4k2. The agent is configured to run daily scraping tasks and has been auto-started. Current status: Active. You can view execution logs and adjust scraping rules in the Agenty dashboard. The agent will begin its first run within the next scheduled interval.
This example shows agent creation with auto-start enabled. Note that you'll need valid Agenty API credentials with agent-creation permissions. The actual scraping configuration (selectors, pagination rules) must be set up separately in the Agenty UI — this MCP handles agent lifecycle, not scraping logic itself.
@agenty clear all rows in list 'daily-leads' then add these agents to project 'Q1-Research': agent_sc_9k3m, agent_pr_2j8n
List 'daily-leads' (ID: list_dl_5x7p) cleared — 847 rows removed. Agents agent_sc_9k3m and agent_pr_2j8n successfully added to project 'Q1-Research' (ID: proj_q1_4h9t). Both agents are now visible in the project dashboard and will contribute data to the project's shared output streams.
This demonstrates chaining destructive and additive operations. Clearing a list is irreversible — all rows are permanently deleted. Ensure you have backups or export data first. Adding agents to a project requires that both the agents and project already exist; the MCP will fail if IDs are invalid.
@agenty list all my API keys, then disable any that haven't been used in the last 30 days to reduce security exposure
Found 7 API keys in your account. Analysis shows 3 keys inactive for 30+ days: key_a8x2 (last used 45 days ago), key_m3k9 (last used 62 days ago), key_p7n4 (last used 91 days ago). These keys have been disabled. The remaining 4 active keys are still enabled. You can re-enable any key via the Agenty dashboard if needed.
This example pairs the MCP's read tools (listing keys) with AI reasoning (filtering by usage date) and write tools (disabling keys). The MCP can toggle key status but doesn't track usage timestamps directly — the AI infers this from metadata. Disabling a key immediately revokes API access, so confirm dependencies before running this workflow.
Use-case deep-dives
When Agenty wins for managing ephemeral API keys at scale
A 6-person data ops team runs 20+ scraping agents that hit rate-limited APIs every hour. They rotate keys weekly to stay under vendor TOS limits and avoid bans. Agenty's key-creation and status-toggle tools let them script the rotation inside Switchy: one prompt generates a fresh key, swaps it into the active agent config, and disables the old one. This works cleanly if your team already uses Agenty for scraping orchestration and you need programmatic key hygiene. If you're just running a handful of static scrapers, the 40-tool surface is overkill—stick with manual key rotation in your vendor dashboard. Buy Agenty integration when credential churn is a weekly tax and you want to automate it in the same workspace where you triage scraper failures.
When this MCP handles rapid agent experimentation
A 3-person growth team tests landing-page scraper configs to find which CSS selectors survive site redesigns. They clone a baseline agent, tweak one selector, run both in parallel, then compare output quality. Agenty's clone-agent tool makes this a one-line Switchy prompt instead of clicking through the vendor UI five times per test. The win is speed: you can spin up 10 variants in a standup and kill the losers by lunch. The threshold is complexity—if your agents have fewer than 3 config parameters, cloning by hand is faster than writing the prompt. This MCP pays off when you're iterating on scraper logic daily and the UI becomes a bottleneck. Use it when agent experimentation is a core loop, not a monthly chore.
When Agenty's list tools beat manual CSV uploads
A 4-person sales ops team syncs Agenty's scraped contact lists into HubSpot every Monday. They clear stale rows, re-populate from fresh scrapes, then export to CSV for upload. Agenty's clear-list and create-list tools let them script the refresh in Switchy: one prompt wipes last week's data, another triggers the scraper run, and a third confirms row counts before export. This saves 20 minutes per sync if you're already using Agenty lists as a staging layer. If your scrapers write directly to your CRM or you sync fewer than 500 rows, the overhead isn't worth it—just download the CSV from Agenty's UI. Buy this integration when list hygiene is a recurring manual step and you want to collapse it into a single Switchy command.
Frequently asked
What does the Agenty MCP do in Switchy?
It lets your AI agents create, clone, and manage Agenty web scraping agents directly from chat. You can spin up new scrapers, add them to projects, clear list data, and toggle API keys without opening the Agenty dashboard. Useful if your team runs dozens of scraping jobs and wants AI to orchestrate them based on conversation context.
Do I need an Agenty API key to connect this MCP?
Yes. You'll paste your Agenty API key into Switchy's connection form. Agenty uses simple API key auth, so there's no OAuth redirect or admin-approval step. Grab the key from your Agenty account settings, and anyone on your Switchy workspace with the key can connect the MCP.
Can the MCP run an Agenty scraper and return the scraped data?
No. The MCP creates and configures agents but doesn't execute scraping runs or fetch result rows. You'll still trigger jobs in Agenty's UI or via their separate run API. Think of this MCP as the control plane for agent setup, not the data pipeline itself.
How is this different from calling Agenty's API directly?
The MCP wraps 40 Agenty API endpoints so your AI can invoke them by name in natural language. Instead of writing curl commands or Python scripts, you say "clone agent xyz and add it to project 123" and the AI translates that into the right sequence of API calls. Faster for ad-hoc tasks; overkill if you already have scripts.
Who on the team should connect the Agenty MCP?
Whoever manages your Agenty account and has an API key with full permissions. Since the MCP can create, clone, and delete agents, you want someone who understands your scraping workflows. Once connected, any Switchy user in the workspace can ask the AI to use it, so set workspace permissions accordingly.