Privacy

Pliure is a personal deployment, not a service

Pliure is software that one person runs for their own data and their own AI agents. There is no public product, no sign-up, no user accounts and no shared infrastructure. Each instance is installed, configured and operated by the individual whose data it holds, who is the sole controller of that instance. This policy describes what that software does with data. It is not an offer of service to anyone else.

Because there are no other users, Pliure collects nothing from visitors to this website: no analytics, no cookies, no trackers, no logging of your visit beyond what any web server records to serve a page.

What an instance reads

An operator connects their own accounts to their own instance. Nothing is read until that connection is authorised, and nothing outside the authorised scopes is ever requested.

Where Google accounts are connected, Pliure may read, depending on the scopes the operator granted:

• Gmail: message headers, bodies and attachments, and labels.
• Google Calendar: calendar lists and event details.
• Google Drive: file metadata and, for files the operator chooses to index, their contents.
• Google Contacts: names and contact details.
• Google Chat: spaces and messages.

Other connectors read the equivalent from the sources an operator chooses to connect, such as messaging apps, note taking tools, code hosting and financial aggregators.

What it is used for, and what it is never used for

Data is used for exactly one purpose: to build a searchable index on the operator's own machine, so that the operator and the agents they authorise can find and reason over their own information.

It is not sold. It is not rented or traded. It is not used for advertising or profiling. It is not used to train, fine tune or evaluate any machine learning model. It is not transferred to any third party, except where the operator themselves directs a specific query to a specific service of their choosing.

Google user data and the Limited Use requirements

Pliure's use and transfer of information received from Google APIs to any other app will adhere to the Google API Services User Data Policy, including the Limited Use requirements.

In practice that means Google user data is used only to provide and improve the operator's own features as described above, is never transferred to others except as that policy permits, is never used for advertising, and is never read by a human other than the operator, except where required by law or necessary for a security investigation of the operator's own instance.

Where it is stored

On hardware the operator controls. Pliure is local first by design: the canonical copy of an instance's index lives on the operator's own machine, not in a hosted multi tenant database. Credentials and access tokens are stored on that same machine, with filesystem permissions restricting them to the operating system account that runs the software.

Traffic between an operator's agents and their instance is encrypted in transit.

Who can read it

The operator. Beyond that, only the agents the operator explicitly authorises. Each agent is issued its own credential, scoped to a named subset of the connected sources rather than handed a copy of the operator's tokens, and each call an agent makes is recorded in an audit log the operator can read.

Retention, and how to end it

An instance keeps what it has indexed until the operator removes it. Three things end access, and any of them is enough:

• Revoking Pliure at myaccount.google.com/permissions, which stops all further reading of that Google account immediately.
• Disconnecting a connector in the instance's own configuration, which stops collection for that source.
• Deleting the instance's local store, which destroys the copy that was collected.

Deleting data in Gmail, Drive or any other connected source does not by itself delete the copy an instance already holds. Removing that copy is the third step above.

Children

Pliure is not directed at children and is not made available to them.

Changes

If this policy changes, the effective date at the top of this page changes with it, and the previous wording remains in the public history of the repository this site is built from.

Contact

There is no central operator to write to, because there is no central service. Each instance is run by one person, who is the controller of the data inside it. If an instance holds data about you, the person to ask is the one who runs it, and they are the person who gave you this link.