Connect fonio.ai to Specky.
AI phone agents — turn post-call transcripts into searchable evidence by webhook, then place approved callbacks.
Signal in
Activity
Context
Linked
Next move
Citable
What fonio.ai does
The tool, in plain language.
fonio.ai runs AI phone agents that answer and handle real calls, including the transcript, direction, caller details, and fields your assistant collects.
Why connect it to Specky?
Keep the useful context attached.
Connect fonio.ai when phone conversations contain product truth you need to compare with support and analytics. Specky receives new calls by webhook because fonio.ai does not expose call history for backfill.
- Voice-of-customer calls
- Call transcripts with structured fields
- A confirmed follow-up call after a product signal
The job this connection does
Why would a product team connect fonio.ai?
Turn new activity from fonio.ai into searchable product evidence while it is still fresh.
fonio.ai pushes to Specky in real time
fonio.ai sends new activity to Specky as it happens. There is no backfill: anything from before you connect it stays in fonio.ai. Set up the webhook first, then let the conversations you care about happen.
Know what this connection is for.
A useful integration is clear about its boundary: what Specky can receive, what it returns to your product workflow, and what it deliberately does not do.
You provide
- A completed-call webhook with the transcript and conversation ID
- Optional extracted fields such as plan tier, order number, or churn reason
- An outbound number only when you want Specky to place a callback
Specky gives back
- A searchable voice-of-customer signal linked to the source call
- Structured call fields that can be filtered and cited in product work
- A confirmed, audit-logged callback when you approve one
A real example
A caller says “exports time out on large projects.” The webhook creates a signal with the transcript and plan tier; you can then ask Specky to compare it with support and analytics before deciding what to do.
Questions you can ask in plain English
These are the kinds of questions this connection helps answer. Specky keeps the source attached so the answer can be checked, not just trusted.
Connecting fonio.ai
Here’s the quick version. The setup guide walks through every field and screen.
- 1
Go to app.fonio.ai/api-keys and create a workspace API key
- 2
Paste it above. If you want Specky to place calls too, also enter one of your fonio.ai numbers in international format (+4312345678)
- 3
Open the assistant in fonio.ai → Integrations → add an HTTP Request integration
- 4
Set When it runs to After the call
- 5
Point it at the webhook URL shown on the fonio.ai card in Settings → Integrations, and add the signing secret as an x-specky-webhook-secret header
Before you connect
Workspace API KeyRequired
Create one at app.fonio.ai → API keys.
Outbound Number (optional)Optional
One of your fonio.ai numbers, in international format. Only needed if you want Specky to place outbound calls — leave blank to just ingest call transcripts.
Catch the fonio.ai conversation while it is fresh
Ask about fonio.ai in plain English instead of digging through another tab
See recurring themes alongside signals from the rest of your stack
Carry the useful quote or record straight into a PRD or decision
Keep the original source attached so someone else can check it
Community project
Want an agent to work with fonio.ai directly?
There is an unofficial, MIT-licensed community MCP for fonio.ai's public API. It is separate from this Specky integration, self-hostable, and clearly not affiliated with fonio GmbH. Use it when you want Claude, Cursor, or another MCP client to work with fonio.ai itself; use this integration when you want finished calls to become evidence in your Product Graph.
View the open-source Fonio MCP on GitHubBefore you connect it
How does Specky use my fonio.ai data?
It becomes part of your Product Graph, where you can search it alongside related feedback from the rest of your stack. The source stays attached, and your workspace data is not used to train AI models.
Will Specky import my existing fonio.ai history?
No — and this is worth knowing before you start. fonio.ai doesn't offer an API that lets Specky ask what happened in the past, so data only arrives from the moment you finish connecting it. Set it up before the conversations you want captured, not after.
What do I need before I connect fonio.ai?
You will need Workspace API Key from your own fonio.ai account. The setup guide walks you through each one.
How long will this take?
Usually a few minutes. You create a credential in fonio.ai, paste it into Specky, and you are connected, plus one step to point fonio.ai's webhook at Specky. Step-by-step instructions are in the setup guide.
Can I disconnect fonio.ai later?
Yes — go to Settings → Integrations whenever you want. Disconnecting removes the connection and the data it brought in.
Can I use fonio.ai on my plan?
Every plan can connect integrations. Your plan sets how many can be active at once, and a few providers are limited to Scale. Settings → Usage shows where you stand.
Can Specky place outbound calls through fonio.ai?
Yes. Once connected with a workspace API key (and optionally an outbound number), Specky can trigger fonio.ai to call a customer — useful when a signal needs a follow-up conversation, not just another ticket.
Bring fonio.ai into your product graph
Connect it in a few minutes, then let Specky do the reading.