Manus for AI agents
Manus is an AI task automation platform that lets users create projects, run tasks, and manage files through a single workflow layer. 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 Manus.
Create a new task in a project
Create a task under a project to kick off a piece of automated work.
Check a task's status
Look up a task's current state and output before deciding on next steps, read-only.
Approve before deleting a task
Require approval before deleting a task, since it removes its files and history.
14 Manus actions, graded by risk.
Every Manus action is tagged read, write or destructive, so one policy covers the whole app and new actions inherit the right default.
read
6Look things up. Allowed by default.
- manus.get_fileGet file details
- manus.get_taskGet Task
- manus.list_filesList uploaded files
- manus.list_tasksList tasks
- manus.list_projectsList Projects
- manus.get_webhook_public_keyGet Webhook Public Key
write
5Create and change things. Allow, or ask the user first.
- manus.create_fileCreate File
- manus.create_taskCreate New Task
- manus.update_taskUpdate Task
- manus.create_projectCreate New Project
- manus.create_webhookCreate Webhook
destructive
3Delete, cancel or archive. Ask first, or deny outright.
- manus.delete_fileDelete a file
- manus.delete_taskDelete task
- manus.delete_webhookDelete Webhook
Manus in three steps.
- 01Your users connect ManusThey add their Manus api key on Arc0 Connect, under your brand. It goes straight into the vault.
- 02You set the rulesReads run, writes like “create File” can wait for the user, and “delete a file” can be denied outright.
- 03Any agent can actYour agent calls Manus through the Arc0 SDK or MCP, and so can Claude, ChatGPT and Cursor. Every call lands on the audit log.
await arc0.policies.set('manus', { read: 'allow', write: 'ask', // create_file destructive: 'deny', // delete_file }) # Claude Code: the same connection, one URL $ claude mcp add --transport http arc0 \ https://mcp.arc0.ai/u/u_8f2
How Manus connects.
Users add their Manus 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 Manus 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 Manus from any agent.
Manus and Arc0, answered.
Can I use Manus with Claude, ChatGPT or Cursor?
Yes. Connect Manus 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 Manus actions you allow.
How do users connect Manus?
Users add their Manus 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 Manus actions can my agent take?
14 in total: 6 read, 5 write and 3 destructive, such as “create File”. Your policies decide which of them each agent may call.
Can I stop my agent from deleting things in Manus?
Yes. Actions like “delete a file” 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 Manus too?
Yes. The same Manus connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.
Plug Manus into your agent.
Your users connect Manus once, under your brand. Your agent gets 14 actions behind your policies, with every call on the record.
Free to build · MCP + REST · Audit log on every plan