Settings

Configure your engine options, credentials, styling preferences, and roadmap updates.

AI Provider Settings

Select which language model engine generates code, and save your API keys locally.

Anthropic
claude-3-5-sonnet
OpenAI API
gpt-4o
NVIDIA NIM
nemotron-3-ultra
No Provider
Simulation only
No key saved

No key saved

No key saved

Appearance

Personalize colors, layouts, and typography across the builder console.

Deep Navy (Dark Theme)
Clean Slate (Light Theme)

Preferences

Control streaming speed, code settings, and debugging console flags.

Token Streaming Animation
Generates code line-by-line using a streaming simulation instead of instant dump.
Show Agent Log Console
Renders the logs panel displaying agents communication under the preview block.
Auto Preview Sandbox
Automatically mounts the code inside the live preview sandbox upon code generation complete.
Multi-Agent Mode
Planner → Coder → Reviewer full cycle vs a faster Single-Agent code completion.

Development Roadmap

Check out the planned upgrades and feature releases for the PromithicAI engine.

v1.0
First Stable Release
Multi-agent simulation pipeline, code viewer (Monaco), theme toggle, and localStorage history.
Completed
v1.1
Supabase Database & Firebase Auth
Integrate database persistence for history sync, user login credentials, and user profiles.
Active
v2.0
Custom LLM Keys Integration
Allows sending API queries directly using user API keys for Claude and OpenAI engines. Includes Connect/Disconnect toggle per provider.
Active
v2.0
LangGraph Orchestration & FastAPI
Move the simulation to a true Python backend graph model using stateful LangGraph agents.
Q3 2026
v2.1
MCP Sandbox Servers
Secure execution sandbox to enable filesystem commands, dependency installs and custom code previews.
Q4 2026
v2.2
GitHub Sync & Deployments
Connect repository workspace, push commits automatically, and deploy one-click previews via Vercel.
Future

About PromithicAI

Technical specification details of the current workspace.

Engine version 2.0.0

FastAPI LangGraph multi-agent orchestration backend. Dynamic model selection, NVIDIA NIM, Claude, OpenAI, with simulation fallback.

Stable