Purpose

Look up Finnish phone numbers on Fonecta Caller and return a phone → name map. Fonecta is the primary Finnish directory service; it covers most publicly listed personal and business numbers.

The script uses Playwright to log in, navigate to each number's search page, and extract results from the __NEXT_DATA__ JSON blob embedded in the SSR-rendered HTML. No undocumented API calls are needed beyond what the browser does naturally.

When to Use

Use this skill when you have a list of Finnish phone numbers (starting with +358 or 0) and need to resolve them to names — for example to enrich a WhatsApp participant list, a call log, or any dataset that contains bare phone numbers.

Do not use for non-Finnish numbers; Fonecta only covers Finland.

Prerequisites

Workflow

1. Prepare the phone list

Normalise all numbers to bare E.164 digits without a leading + (e.g. 358401234567). The script accepts any format — +358…, 0…, 358… — and normalises internally, but a clean list avoids surprises.

Write the numbers to a file, one per line, or pass them directly with --phones.

2. Run the lookup

FONECTA_EMAIL=user@example.com FONECTA_PASSWORD=secret \
    uv run scripts/fonecta_lookup.py \
        --phones-file phones.txt \
        --output fonecta-names.yaml

Or with inline numbers:

FONECTA_EMAIL=... FONECTA_PASSWORD=secret \
    uv run scripts/fonecta_lookup.py \
        --phones 358401234567 358407654321 \
        --output fonecta-names.yaml

The script is incremental: if --output already exists, previously resolved numbers are skipped. Reruns after a partial failure only look up the remainder.

3. Read the results

fonecta-names.yaml contains a mapping of normalised E.164 digits → name string (or null if not found):

358401234567: Matti Meikäläinen
358407654321: Yritys Oy
358409999999: null   # not found in Fonecta

Pass this file to whatever downstream process needs the names.

How Results Are Extracted

The search page https://www.fonecta.fi/haku/<local-number> is a Next.js SSR page. The search results are embedded in the page HTML as a __NEXT_DATA__ JSON blob populated by React Query (dehydratedState). The script parses this directly — no additional API call is needed.

The relevant path inside __NEXT_DATA__:

props.pageProps.dehydratedState.queries[]
  → queryKey[0] == "search"
  → state.data.results[0].displayName

See references/fonecta-api.md for full API details, the authentication flow, and notes on rate limiting.

Login Flow Detail

The login form is a two-step MUI Dialog:

1. Accept the OneTrust cookie consent banner.
2. Click the "Kirjaudu sisään" nav button (opens the dialog).
3. Enter email → click "Seuraava".
4. Enter password → click the "Kirjaudu sisään" button scoped to the dialog (the nav-bar button is visually blocked by the dialog overlay and times out if clicked directly).

Known Limitations

• Numbers that are unlisted, prepaid SIMs, or business lines not registered with Fonecta return null. Approximately 25 % of Finnish mobile numbers are not in the directory.
• The fofisuggest autocomplete endpoint consistently returns empty results even for numbers that the full search page resolves; ignore it.
• The script runs headless. If Fonecta introduces bot-detection measures, switching to headless=False may help.
• Session tokens expire after ~30 minutes. For batches longer than ~200 numbers, add token refresh logic (see references/fonecta-api.md).