Spend Control for ElevenLabs
ElevenLabs is the TTS backbone for voice agents — billed per character. The guard caps the voice line.
Why ElevenLabs agents spend
Per-character billing at conversation volume.
Premium voices cost more per character.
How it works
Call the guard before TTS runs: amount, merchant, category → APPROVED, BLOCKED, or FLAGGED.
Rules that fit
Per-agent daily ceiling on the voice category.
Velocity limit on regeneration loops.
Guard call
import requests
r = requests.post("https://sipi.bot/v1/transactions/evaluate",
json={"amount": 4.2, "merchant": "elevenlabs", "category": "voice"},
headers={"Authorization": "Bearer KEY"})
decision = r.json()["decision"]Characters are a bill; the gate makes it a budget.
FAQ
Does this slow TTS?
No — ~5 ms per check.
Can I cap per voice agent?
Yes — per-agent rules per agent.
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