For teams shipping changes to a live voice agent

The prompt text didn’t change.The agent talks over callers now.

A text diff tells you the words are the same. It can’t tell you the agent stopped yielding when callers interrupt, or that its latency tail doubled, or that it quit escalating. This library generates fixed canary calls and scores every release against your own baseline on the behaviors that actually decide a call.

Get the Library — $79one-time · instant download · yours to keep
Five deliverables · runnable
Runnable Python generator + scorer
stdlib
Workbook that reproduces it
xlsx
Release-score export
csv
Canary-design playbook
docx
Regression-triage runbook
docx
Behavior layer beside
Prompt Regression Lab (text) · Voice Prompt Forge (build)
01.The Problem

A green text diff hides a voice regression.

5 → 4
one dropped barge-in stop: the agent now talks over callers, words unchanged
950 → 1250ms
a latency tail that doubled while the transcript looked identical
your run
graded against the baseline you shipped, not an industry number
02.See It Work

Same correct answer. Drop the barge-in. Watch it break.

This release vs baseline · 4 canaries
CN-001Booking · hours
ON BASELINE
CN-002Booking · appointment · latency 1250ms > baseline 950ms + slack
BELOW BASELINE
CN-004Pricing question
ON BASELINE
CN-007Escalation demand
ON BASELINE
3 on · 1 below · 0 brokeSLIPS TO REVIEW

CN-002 is already below baseline in both states — same correct answer, but p95 latency jumped from 950ms to 1250ms. Correctness held; the call got slower. The library grades the behavior, not just the words.

03.The Runnable Scorer

Generate the library once. Score every release against baseline.

A zero-dependency Python tool generates the canary library from your topics and scores a run against your baseline. Here is its actual score output on the shipped seven-canary sample — a release with two slips and two breaks:

CANARY CALL LIBRARY - RELEASE SCORE
====================================================

PER-CANARY (this run vs baseline)
----------------------------------------------------
CN-001   ON BASELINE      
CN-002   BELOW BASELINE   latency_p95_ms 1250ms > baseline 950ms + slack
CN-003   ON BASELINE      
CN-004   CANARY BROKE     barge_in 4 < floor 5
CN-005   BELOW BASELINE   dead_air 2 > baseline 0
CN-006   ON BASELINE      
CN-007   CANARY BROKE     escalation 0 < floor 1

ROLLUP
----------------------------------------------------
Canaries scored .......... 7
On baseline .............. 3
Below baseline ........... 2
Canary broke ............. 2
Review first ............. CN-004

VERDICT: REGRESSIONS FOUND

Scores a run against your own baseline run. It places no calls and
grades no person. Not legal advice.
04.The Standard

Three rules the library holds every release to.

Same calls, every release

A canary is only a signal if it never changes. The generator builds a fixed library from your topics — hours, booking, a grounding trap, a barge-in test, an escalation demand — so any difference between runs is your agent, not the test.

Your baseline, not a benchmark

A regression is your agent getting worse than the run you were happy to ship. Every later run is scored against that baseline with tolerance — industry latency numbers are context, not a pass mark.

Some drills don’t bend

Barge-in below 5 of 5, or a scripted escalation that didn’t fire, forces CANARY BROKE regardless of how correct the answers were. An agent that talks over callers fails the call — the words being right doesn’t save it.

05.What This Is — And Isn’t

A voice-behavior regression check. Not a text diff, not a dialer.

It is
  • A generator for a fixed canary-call library from your topics.
  • A scorer of voice behavior — barge-in, latency, dead-air, escalation.
  • Baseline-relative, with a dispositive barge-in / escalation gate.
  • Deterministic — the same two runs always score the same.
It isn’t
  • A prompt-text diff — that’s the Prompt Regression Lab.
  • A dialer or call simulator — it scores results you record.
  • A benchmark against other companies’ numbers.
  • A judge of people — it grades runs, never person.

Scope note. The Canary Call Library scores runs you record against your own baseline; it places no calls, connects to nothing, and grades no person. It is a release-quality aid, not a certification, and not legal advice.

06.Who It’s For

Anyone who ships changes to a voice agent that’s already live.

Agencies pushing prompt or model updates across client voice agents
Operators who changed one thing and want to know what else moved
Teams adding a CI gate before a voice release goes out
Anyone who’s shipped a ‘small tweak’ that quietly broke barge-in
08.Common Questions

The parts worth pinning down before you buy.

The Prompt Regression Lab diffs prompt text — did the output change, ship / hold / regressed. This scores voice behavior a text diff can't see: a release where the agent still says the right words but now talks over callers, or its p95 latency doubled, or it stopped firing a scripted escalation. Same job (catch a regression) at a different layer — the two are complements, not substitutes. One watches the words; this watches the way it handles the call.

Ship the change.
Know what else it moved.

Run the same canary calls every release and score them against the version you trusted. One purchase, lifetime access, 12 months of updates. $79, once.

Sold by RedHub AI LLC · Secured by Stripe · redhub.ai