CypherMonkeyAI omni-channel lab

A lifelong-learner lab on AI omni-channel agents, including voice and the contact center. Free to read. Nothing sold.

Voice and omni-channel agents · L2 · 15 min

Chat and messaging beside voice

Learn in five minutes

  • Chat can show the quote on screen. Voice has to say it.
  • A message can arrive later. The person may not be waiting.
  • The same refusal text should not promise a callback the notes do not allow.
  • Identity is weaker on a new message thread until you have checked it.
  • Do not copy a vendor's channel list as if it were complete.

The voice reply is one sentence plus the offer to read the quote in chat. The chat reply shows the quote. The policy is the same.

This section HTML

Visual

T7-R4 diagram

Diagram SVGDiagram PNGDiagram HTML

Two wordings

  1. Policy: quote or refuse.
  2. Voice wording: one sentence and an offer to show the quote.
  3. Chat wording: the quote on screen.
  4. Neither wording adds a refund the note lacks.

This section HTML

Quiz

What can chat show that voice must speak?

The channel changes the presentation.

What is different about a later message?

Waiting changes.

What must both wordings avoid?

The note is still the source.

This section HTML

Ecosystem

Wording
The channel-specific sentence. The policy stays put.

Sisters: Web Speech API, Turn-taking, Queues, CRM notes.

The retrieval index and the tool cap are shared. The channel changes the wait and the consent copy.

A later host could add a channel without rewriting the notes server.

Config

  • Write the goal in one sentence.
  • Name the input and the allowed output.
  • Save the trace next to the result.

Troubleshooting

  • The answer looks fluent and is wrong. Check whether a source passage actually contains the claim.
  • The loop never stops. Set a step cap and stop when the tool result repeats.

This section HTML

Sources HTML

Sources and go deeper

sources.json