Antigravity Gateway • Console

Antigravity Harness v1.2.0

127.0.0.1:8045 • IDE Linked

Active: Detecting...

Next Reset: Calculating...

Next in Line: Calculating...

Shield: Armed (<20%)

Tailnet: Connecting...

Pool Headroom: Healthy

Pool Aggregate Headroom --%

~18.5M tokens available 4/4 accounts active

Velocity & Burn Rate

-- tokens/hr

Today: -- tokens burned

Active Session Runway Healthy

Calculating runway...

Next failover: Calculating...

Quota Replenishment Chrono-Timeline 0 Upcoming

5h Burst Weekly Reset Live Cloud Code Feed

Loading reset timelines...

	KEY	STATUS	ACCOUNT & ALIAS	GEMINI PRO (WEEKLY)	5H BURST HEADROOM	CLAUDE / 3P QUOTA	USAGE STATS	ACTION
Syncing live account pool telemetry...

Smart Quota Shield Configuration • Automatic proactive failover between tasks without breaking conversations

Failover Thresholds

Proactive Engine

Switch When Weekly Below Proactive failover safety buffer for 7-day model quota

Switch When 5-Hour Burst Below Proactive switch buffer for fast 5-hour burst window

Auto-Switch Protection Enable background proactive account rotation when limits approach

Model & Account Policies

Intelligent Routing

Watch Limits Of Auto-detect from conversation model

Main Account (Affinity) Preferred home account; returns once quota recovers

Auto-Continue After Limit Send 'continue' automatically after hard quota switch

Claude / 3P Routing Strategy Smart preservation & failover for Anthropic models

Smart Shield Failover & Routing Simulator Zero IDE Impact

Simulate an unexpected 429 quota exhaustion or weekly limit breach on your active account to dry-run how the Smart Shield routing algorithm evaluates headroom, urgency, and candidate scores in real-time.

Token Telemetry & API Cost Savings • Calculated against standard Google Cloud Pro rates

Total Tokens Processed

Across all conversations & agents

Input / Context Tokens

Prompts, files, system instructions

Output Completion Tokens

Agent tool calls and completions

Commercial API Value Saved

$0.00

Cumulative free quota utilized

14-Day Token Velocity & Daily Savings

Daily token throughput from local agent sessions

$0.00 Saved Today

Model Distribution

7-Day Hourly Quota Burn Heatmap

Hourly request density & token throughput matrix across all local agents

Peak: -- tokens

Active coding hours tracked in SQLite

Less

Enterprise ROI & Quota Savings Analysis

Financial comparison vs commercial API pay-per-token rates (OpenAI / Claude standard list prices)

Commercial Value $0.00 Standard API meter rate

Antigravity Cost $0.00 100% Free Quota Pooled

Net Dollars Kept $0.00 Total savings delivered

Live Dispatch Stream • Real-time event notifications via SSE

[INIT] [HARNESS] Gateway active on 127.0.0.1:8045. Multi-account routing armed.

Filtered Requests 0

Success Rate 100%

Avg Latency 0ms

Tokens Processed 0

STATUS	REQ ID	ACCOUNT	MODEL	LATENCY	TOKENS	TIMESTAMP	INSPECT

Showing 0 of 0 requests

Interactive Model Playground • Test queries directly against the load balancer and inspect live token deduction

Model

Quick Presets

Prompt

⌘ + Enter to run

Response Telemetry

Ready

Served By

Latency

-- ms

Tokens

-- / --

Cost Saved

$0.00

Click "Send Request" to test model generation and load balancing through the gateway.

Interactive Quota Simulation & What-If Sandbox

• Model multi-agent workloads, forecast cluster exhaustion, and visualize failover cascades

Presets:

Workload Parameters

Concurrent Agents / Engineers 3 agents

1 agent 10 agents 25 agents

Average Prompt & Context Size 20,000 tokens

2k (Quick edits) 30k (Large files) 100k (Whole repo)

Turns per Hour (per Agent) 15 turns/hr

Low (5/hr) Active (20/hr) Autonomous (60/hr)

Primary Model

Forecast Window

💡 Smart Shield Protection Rule: Accounts automatically failover when remaining burst headroom dips below the configured 20% threshold.

Hourly Token Burn

900,000

tokens / hour

Window Demand

4.5M

in 5h burst window

Commercial API Cost

$31.50

Value saved via harness

Sequential Account Failover Cascade

Predictive Smart Shield Dispatch

Universal API & Protocol Translator • Use Antigravity's pooled Pro accounts in any OpenAI or Anthropic tool

Tailscale Remote AI Mesh Online

Encrypted WireGuard Mesh • MagicDNS Active

MagicDNS Secure URL

https://md.tail58fb78.ts.net

Cursor / VS Code Base URL

https://md.tail58fb78.ts.net/v1

Gateway Virtual API Keys 0 Keys

Create and manage Bearer API keys to secure remote Tailscale & public access

No API keys created yet. Without an API key, external access is unrestricted. Create a key to protect your Gateway on public internet.

OpenAI Compatible Endpoint

Compatible with Cursor, Cline, Continue, Aider, and standard OpenAI SDKs.

curl http://127.0.0.1:8045/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gemini-3.8-flash",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

Anthropic Compatible Endpoint

Direct protocol support for Anthropic Messages API.

curl http://127.0.0.1:8045/v1/messages \
  -H "Content-Type: application/json" \
  -d '{
    "model": "claude-3-5-sonnet",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

Live Gateway Test Console

Send a live test completion through the active account pool