Spend Control for Roo Code
Roo Code (formerly Roo Cline) runs autonomous coding agents in the editor — capable of buying tools and compute mid-task. The guard keeps the budget.
Why Roo Code agents overspend
Long autonomous sessions call paid tools repeatedly.
Task-planning loops can trigger purchases.
No built-in dollar-level budget.
How it works
Attach the guard as a tool. Before any spend, the agent asks sipi.bot and gets APPROVED, BLOCKED, or FLAGGED.
Rules that fit Roo Code workflows
Per-session or per-agent daily ceiling.
Merchant allowlist for paid tools.
Velocity limit on retry loops.
Guard call
const res = await fetch("https://sipi.bot/v1/transactions/evaluate", {
method: "POST", headers: { "Content-Type": "application/json" },
body: JSON.stringify({ amount: 30, merchant: "api.provider.com", category: "api" }),
});
const { decision } = await res.json();One tool, enforced budget.
FAQ
Can I cap per project?
Yes — per-agent rules per project or workspace.
Is the decision model-free?
Yes — deterministic rules, ~5 ms.
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