Apify for AI agents
Apify is a platform for building, running, and managing web scraping and automation programs called Actors, used by developers and data teams to pull structured data from the web at scale. Connect it once through Arc0, and your agent, or Claude, ChatGPT and Cursor, can use it through one MCP endpoint, limited to what each user approved.
What agents do in Apify.
Run a scraping actor and get results
Run a task synchronously and retrieve the resulting dataset items right away.
Check the status of an actor run
Look up an actor run or build to see whether it finished successfully.
Confirm before aborting a run
Stopping an in-progress actor run should be confirmed since it may lose partial results.
183 Apify actions, graded by risk.
Every Apify action is tagged read, write or destructive, so one policy covers the whole app and new actions inherit the right default.
read
86Look things up. Allowed by default.
- apify.get_logGet log
- apify.get_actorGet Actor Details
- apify.search_googleSearch Google
- apify.get_task_inputGet Task Input
- apify.list_user_runsList User Actor Runs
- apify.get_all_webhooksGet all webhooks
- apify.get_list_of_runsGet list of runs
- apify.search_instagramSearch Instagram
- apify.get_dataset_itemsGet dataset items
- apify.get_default_buildGet Default Build
- apify.get_list_of_tasksGet list of tasks
- apify.get_list_of_buildsGet list of builds
- apify.search_google_newsSearch Google News
- apify.search_upwork_jobsSearch Upwork Jobs
- apify.get_key_value_recordGet Key-Value Record
- apify.search_google_imagesSearch Google Images
write
82Create and change things. Allow, or ask the user first.
- apify.create_taskCreate Actor Task
- apify.create_actorCreate Actor
- apify.create_datasetCreate Dataset
- apify.update_task_inputUpdate Task Input
- apify.create_task_webhookCreate Task Webhook
- apify.update_key_value_storeUpdate Key-Value Store
- apify.act_putUpdate Actor
- apify.acts_getGet list of Actors
- apify.run_taskRun Task Asynchronously
- apify.user_getGet Public User Data
- apify.run_actorRun Actor Asynchronously
- apify.store_getGet list of Actors in Store
- apify.dataset_getGet Dataset
- apify.dataset_putUpdate Dataset
- apify.webhook_getGet webhook
- apify.webhook_putUpdate Webhook
destructive
15Delete, cancel or archive. Ask first, or deny outright.
- apify.delete_actorDelete Actor
- apify.delete_webhookDelete Webhook
- apify.dataset_deleteDelete Dataset
- apify.schedule_deleteDelete Schedule
- apify.actor_run_deleteDelete Actor Run
- apify.actor_task_deleteDelete Actor Task
- apify.actor_build_deleteDelete Actor Build
- apify.act_version_deleteDelete Actor Version
- apify.request_queue_deleteDelete Request Queue
- apify.key_value_store_deleteDelete Key-Value Store
- apify.act_version_env_var_deleteDelete Actor Version Environment Variable
- apify.request_queue_request_deleteDelete Request from Queue
- apify.key_value_store_record_deleteDelete Key-Value Store Record
- apify.request_queue_request_lock_deleteDelete Request Lock
- apify.request_queue_requests_batch_deleteBatch Delete Requests from Queue
Apify in three steps.
- 01Your users connect ApifyThey add their Apify api key on Arc0 Connect, under your brand. It goes straight into the vault.
- 02You set the rulesReads run, writes like “create Actor Task” can wait for the user, and “delete Actor” can be denied outright.
- 03Any agent can actYour agent calls Apify through the Arc0 SDK or MCP, and so can Claude, ChatGPT and Cursor. Every call lands on the audit log.
await arc0.policies.set('apify', { read: 'allow', write: 'ask', // create_task destructive: 'deny', // delete_actor }) # Claude Code: the same connection, one URL $ claude mcp add --transport http arc0 \ https://mcp.arc0.ai/u/u_8f2
How Apify connects.
Users add their Apify api key on Arc0 Connect. It is encrypted in the vault, never shown to the model, and each user can rotate or revoke it at any time.
The same Apify connection serves your agent over MCP and your own backend over REST and the proxy, so a user connects once. How Arc0 handles credentials →
- AUTH
- API key
- CREDENTIALS
- Per-tenant encrypted vault
- MODEL SEES
- Results only, never credentials
- AUDIT LOG
- Every call, on every plan
Use Apify from any agent.
More web data and scraping apps.
Apify and Arc0, answered.
Can I use Apify with Claude, ChatGPT or Cursor?
Yes. Connect Apify to Arc0 once, then add your Arc0 MCP URL to Claude, ChatGPT, Cursor, Claude Code or any other remote-MCP client. Each assistant only gets the Apify actions you allow.
How do users connect Apify?
Users add their Apify api key on Arc0 Connect. It is encrypted in the vault, never shown to the model, and each user can rotate or revoke it at any time.
Which Apify actions can my agent take?
183 in total: 86 read, 82 write and 15 destructive, such as “create Actor Task”. Your policies decide which of them each agent may call.
Can I stop my agent from deleting things in Apify?
Yes. Actions like “delete Actor” are graded destructive. Set destructive actions to deny, or to ask so the user approves each one, and blocked calls still show up on the audit log.
Can my own backend call Apify too?
Yes. The same Apify connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.
Plug Apify into your agent.
Your users connect Apify once, under your brand. Your agent gets 183 actions behind your policies, with every call on the record.
Free to build · MCP + REST · Audit log on every plan