Gernonimo AI Everywhere

Geronimo — AI built into every PolySaaS admin page
PolySaaS Online — Feature Overview
/admin/geronimo/

The chat box that already knows
which admin page you’re on.

Geronimo sits on every screen in the PolySaaS admin interface. It sees the page you’re working from, answers with that context built in, and can run research through any LLM or through your own Django models.

Every admin pageNot a separate tool to open
Any LLMSwap models per tenant or per task
Two layoutsOverlay or full-width panel
Geronimo
You

What can I configure from here?

Geronimo

You’re on the Slack passthrough page for PolySaaSOnline. From here you can rename the app, restrict it by role, and wire it into an orchestration action.

context: /dose/apps/slack/
Message… (Enter to send, Shift+Enter for newline)
Send

Built around the page you’re already on

Geronimo doesn’t wait to be told what you’re looking at. It reads the admin context automatically and shapes every answer around it.

CONTEXT

Page-aware by default

Ask it “what can I do here” on any screen and it answers for that screen — the passthrough app, the dashboard, the admin panel, whatever’s open.

RESEARCH

LLM lookups on demand

When a question needs outside knowledge, Geronimo can run the research through its configured model and bring the answer back into the same thread.

DATA

Reads your own models

Beyond outside research, it can query the Django ML and application components behind PolySaaS to answer with your actual data.

Point it at any model

Each Geronimo instance is configured independently. Change the underlying LLM per tenant, per team, or per admin section — without touching the interface.

Claude Sonnet anthropic / claude-sonnet-4-6
GPT-4 openai / gpt-4
Local model self-hosted / llama-70b
PolySaaS ML component django-native

Configuration lives at the admin level, not the code level.

  • Assign a different provider to different tenants without a deploy
  • Fall back to a self-hosted model where data can’t leave the network
  • Route a question straight to a Django ML component instead of an external LLM when that’s the faster, more accurate path

Overlay when you need it, panel when you don’t

Geronimo can float over the page for quick questions, or expand into a full-width panel for longer working sessions. Try switching below.

Geronimo

Overlay: stays out of the way of the page underneath — best for a quick check mid-task.

Available everywhere in the shell

One Geronimo instance, present across the interface — not a plugin you add to individual apps.

Passthrough apps
Admin panel
Dashboard
Orchestration bar

Turn it on for your tenant

Geronimo is configured from the admin panel — pick a model, set a display mode, and it’s live across every page your team already uses.

PolySaaS Online Geronimo AI · /dose/api/passthroughendpoints/