Permissions & access
You decide what your assistant can access and when it can act on your behalf.
Perisclaw only ever does what you allow. You decide which services it connects to, which chats it can see, and what it can do in each. You can change any of it at any time from Settings → Permissions in the web app.
What your agent can access
Your agent only reaches what you connect and authorize. Depending on your setup, that can include:
- WhatsApp: your chats, contacts, groups, media, and message history.
- Calendar: events, attendees, and availability on Google Calendar or Outlook.
- Email: your Gmail or Outlook mailbox once you connect it.
- Files: documents in Google Drive and Sheets, or OneDrive.
- Contacts: your Google address book, if you connect it.
- Other tools: anything you explicitly connect through a connector.
It uses this only to carry out what you ask: drafting a reply, scheduling a meeting, tracking a follow-up. See Connect your apps to manage what is linked.
What it can read
Your agent can see your chats in order to do its work. That is what makes it useful. Any chat you block is the exception: it cannot read, search, summarize, or quote a blocked chat at all. To block one, search for it under Blocked chats in Settings → Permissions, choose Block, and confirm, or say “never read <chat>” in your self-chat. Blocked chats are listed on the same page; choose Unblock and confirm to let your agent read one again. Your own chat with Perisclaw can't be blocked.
When it can act
Nothing goes out on its own unless you have set it up. Your self-chat always
works, and so does /claw in any chat. Everywhere else, your agent acts only
through work you have created, such as an Automation or a
Mission.
To have it answer in a particular chat, say so in your self-chat: which chat,
what for, and for how long. For example, “answer product questions in the
Launch group for the next 2 hours.” Perisclaw sets that up as an Automation
you can review, pause, or delete. To stop a reply that is already under way,
send /claw stop in that chat; to stop it for good, pause the Automation.
Answer product questions in the Launch group for the next 2 hours.
Pause all sending
Say “pause everything” or “don't message anyone until I say” in your self-chat. Until you say to resume, Perisclaw sends nothing to any chat except your self-chat. That covers replies, Automations, Missions, scheduled messages, and connected AI clients. Messages you send yourself, including from the reply box in the web app, still go out.
Your Automations and Missions stay set up and carry on after you resume. A message that was held while paused is not sent later.
Hard rules
Hard rules are instructions your agent checks before every action and every reply, in every chat, such as “never send a payment link without asking me.” Edit them under Hard rules in Settings → Permissions, one rule per line.
How a message reaches another chat
- You ask it to. It drafts the message, shows it in your self-chat, and sends once you react 👍. If you say “send it now,” it sends straight away. For a chat you message often, say “send to Ravi without drafts” and later messages you ask it to send there skip the draft. Say “always draft for Ravi again” to undo it.
- You invoke it with
/claw. It responds directly in the chat where you called it, labelled as Perisclaw. You are standing in that chat, so there is nothing to preview. - Work you set up runs. Automations and Missions message people as you, within the purpose you gave them.
Everything it sends for you goes out as you, with no assistant label;
only your self-chat and /claw replies are labelled as Perisclaw. To send a
single message labelled as Perisclaw, ask for it: “send this to the group as
Perisclaw.”
It confirms before it acts
By default, anything it sends or changes on your behalf is shown to you for approval first. When you want it to run longer work without checking in at every step, you set that up deliberately as an Automation or a Mission. Even then it never commits money, agrees to a price, or makes a binding promise without checking with you.
Actions that always stop for youHard limits that no setting removes
A few actions are capped no matter what you have allowed, because they cannot be taken back:
Sending the same message to many chats. It stops well before anything looks like a broadcast, reports exactly who has and has not been messaged, and warns that bulk outreach can get your number flagged by Meta. Splitting a broadcast across turns does not work, by design.
Deleting chats or messages in bulk. Hard limits per turn, with a precise report before asking to continue.
Adding people to groups en masse. Capped at 10 people at a time and 20 in one turn. For more, it offers the group invite link, which people join by choice.
Anything running unattended (an Automation or a Mission) gets no confirmation path at these limits. It stops and brings the decision to you.
What it won't say in someone else's chatYour private context stays private
When your agent acts inside another person’s chat, it treats them as an outsider. It will not disclose your other conversations, calendar, email, files, or notes to them, even if they ask directly. If someone probes for your private information, it declines and tells you about it instead.
Manage your data and connections
Connect a service when you want the capability and disconnect it the moment you do not; disconnecting immediately stops all future access. You can also delete specific data, clear what your agent remembers, or delete your account entirely. See Account & settings.
For how your data is stored and protected, see Privacy & security.