Spend Control for Vertex AI Agents
Vertex AI agents run on Gemini with tool access — and tools spend. The guard makes the budget real on Google's platform.
Why Vertex agents overspend
Gemini token costs per turn at agentic volume.
Tool calls hit paid APIs and services mid-task.
GCP budgets cover Google; the rest is ungoverned.
How it works
Add the guard as a tool (or call the HTTP API from a custom tool). Before any spend, the agent gets a deterministic decision.
Rules that fit
Per-agent daily ceiling.
Merchant allowlist for paid tools.
Category rule: Gemini vs tool spend.
Guard call
from sipi_guard import sipi_guard
decision = sipi_guard(amount=35, merchant="data-api.com", category="data")
# APPROVED | BLOCKED | FLAGGEDGemini for the model; the firewall for the money.
FAQ
Does this replace GCP budgets?
No — keep GCP budgets for Google. The firewall covers every merchant.
Is there a language SDK?
sipi.bot is HTTP — any language works.
Related
Stop the next $12,400 night.
One API call (or MCP tool) in front of every agent transaction — APPROVED, BLOCKED, or FLAGGED, deterministic, ~5 ms, fully logged.
See plans — from $99/mo Try a live check