DEDUPED Hot AI repos, turned into use cases Get the Monday 5

How to manage several agent sessions in a browser GUI with Nosis

Run the GUI to keep multiple sessions, attachments, memory, and scheduled tasks in one place.

Credentials. Attachments and provider keys are stored on disk under ~/.nosis/, so review what you upload and where the keys live.

Who it is forA knowledge worker who juggles several ongoing agent tasks and wants a visual file tree and attachment support.

The problemTerminal sessions are hard to track when several tasks run at once, and files and reminders are scattered across tools.

The outcomeA running GUI where you can open multiple sessions, attach files, browse the workspace file tree, and review memory and scheduled tasks.

What you need first

  • Python 3.11 or newer
  • Node.js 22 or newer
  • uv installed
  • An API key for the default provider, such as OPENAI_KEY

Steps

  1. Complete the install steps for Nosis, including uv tool install --editable ".[gui]".
  2. Run nosis-gui in a terminal and leave it running.
  3. Open the GUI in your browser at the address the nosis-gui process reports, since the README does not list a fixed port. (inferred, check the README)
  4. Create a session, attach a file, and send a request that uses it.
  5. Open the memory and scheduled task views to review what the agent stored and when it will run next.

Prompt for your agent

Using the Nosis repo at https://github.com/EvannZhongg/Nosis, start `nosis-gui` and confirm the GUI loads in a browser. Create two sessions for <task name> and report the session list and the workspace file tree. Verify that a session's history is still present after you restart `nosis-gui`.

Caveat. The README leaves the GUI listen address blank, so confirm the address from the process output rather than assuming a default port.

Share: X LinkedIn Hacker News Email Copy link

Third-party repos we don't own. AI-written use cases can be wrong. Disclaimer