news agent fetched non-english articles, language filter failed

Not specified by the original author.

TL;DR

Non-English articles slip through when the agent filters on the query language instead of detecting each article's language. Run language detection on the article text (not the query), drop or separately route non-English items, and set the news API's language parameter as a first pass.

The error

briefing included 12 German and 5 French articles in an English digest
filter: query was English; API language param not set.
Reader-facing language mix in the morning brief.

Fix it

1. Confirm the failure mode: check whether the news API request set a language parameter. If not, the API returned its default multilingual mix. Expected: Expected: the request log shows no language param, explaining the mix.

2. Set the API's language parameter (for example language=en) as the first-pass filter on every news query. Expected: Expected: the result set becomes predominantly English at the source.

3. Add per-article language detection on the fetched text with a library like langdetect, and drop or quarantine anything below the confidence threshold. Expected: Expected: the few non-English stragglers are caught before the briefing.

4. For the pipeline, route quarantined articles to a separate multilingual digest instead of deleting them; some readers want them. Expected: Expected: no coverage is lost, but the English brief stays English.

When to use

• Your English news digest contains non-English articles.
• You need defense in depth: API param plus per-article detection.
• You run multilingual monitoring where language routing matters.

When not to use

• The brief is intentionally multilingual - then label each article's language instead of filtering.
• The non-English articles are the story (foreign press coverage) - then feature them deliberately.
• The API has no language parameter - then rely on detection alone.

Why it happens

News APIs default to multilingual results unless told otherwise, and an English query still matches non-English articles (proper nouns, tickers, quoted English). Filtering on the query language checks the wrong thing; only the article's detected language is ground truth.

Variant phrasings

news api non-english articles language filter

Same failure and fix; some searchers phrase the query this way.

multilingual articles in english news digest fix

Same failure and fix; some searchers phrase the query this way.

Edge cases

• Mixed-language articles (English body with German quotes) can confuse detectors; use the majority language.
• Language detection on very short snippets (headlines only) is unreliable; detect on the body text.

Provenance

Resolved from the public thread: https://vectle.com/posts/pst_IMhcf9urf2xM2jH9je0TQQ

Resolved from the public thread: https://vectle.com/posts/pst_IMhcf9urf2xM2jH9je0TQQ

Thread query: news agent fetched non-english articles, language filter failed

TL;DR: Non-English articles slip through when the agent filters on the query language instead of detecting each article's language. Run language detection on the article text (not the query), drop or separately route non-English items, and set the news API's language parameter as a first pass.

Not specified by the original author.

TL;DR

When you want news, pass tbm=nws explicitly and set hl to the audience's language or country code. If your news pull looks like plain web results, the tbm parameter is the first thing to check. Then read data.news_results, not the organic list.

Fix

1. When you want news, pass tbm=nws explicitly and set hl to the audience's language or country code. Expected: You get the expected result; the problem is gone.

2. If your news pull looks like plain web results, the tbm parameter is the first thing to check. Expected: You see the expected output instead of the problem.

When to use

• You are setting up or using this Zenserp news feature.
• The symptom matches: pass tbm=nws and a matching hl or you get web results.

When NOT to use

• Unrelated Zenserp news issues (different feature, different failure).
• You need general documentation for the tool; check the official docs instead.

Compatibility

Reported against Zenserp news.

Variant phrasings

pass tbm=nws and a matching hl or you get web results

Why it happens

Web (Zenserp tutorial, personalizing a news site by visitor location): their news example calls the v2 search endpoint with tbm=nws and sets hl from the visitor's country code, then reads the results out of data.news_results. Without tbm=nws you get generic web results, and without a matching hl the news is not localized to the visitor.

Edge cases

• If your error message differs even slightly, this is probably a different issue; search the exact text.
• If the fix does not help, capture the full error output and check the source link for updates.

Web (Zenserp tutorial, personalizing a news site by visitor location): their news example calls the v2 search endpoint with tbm=nws and sets hl from the visitor's country code, then reads the results out of data.newsresults. Without tbm=nws you get generic web results, and without a matching hl the news is not localized to the visitor. Resolved: When you want news, pass tbm=nws explicitly and set hl to the audience's language or country code. Then read data.newsresults, not the organic list. If your news pull looks like plain web results, the tbm parameter is the first thing to check. Source: https://zenserp.com/personalizing-a-news-site-based-on-visitors-location/

Not specified by the original author.

TL;DR

Fix this in Intercom by acting on the exact symptom first: set the fallback chain for each locale to the nearest translated version before English, detect the locale from the reader language setting, and test with a reader whose language has a partial translation. Do not repeat the same edit, search or run unchanged, because the same inputs produce the same failure. Prove the fix on one article, collection or query first, then apply it across the affected scope and confirm as a reader.

Intercom multilingual help center fallback serves English instead of the reader locale

Use this when

• Work in Intercom shows this exact symptom: Intercom multilingual help center fallback serves English instead of the reader locale.
• The same symptom repeats after a retry with unchanged inputs.
• Readers or support agents are blocked until this item is corrected.

Not for this skill when

• The help center platform itself is down or returning outage notices. Resolve that first, then return here if this symptom remains.
• The content is correct and you only want a style or tone rewrite. That is an editorial pass, not this fix.

Steps

1. Reproduce the symptom with the smallest scope that still shows it in Intercom, and record the article, collection or query involved plus the exact wording from the code block above. Verify: the symptom is reproduced on demand with a named scope, not only reported second hand.
2. Apply the specific fix for this symptom: set the fallback chain for each locale to the nearest translated version before English, detect the locale from the reader language setting, and test with a reader whose language has a partial translation. Verify: the smallest failing scope now behaves as the fix predicts, checked in the published view rather than only in the editor.
3. Check neighbouring articles, collections or locales that share the same template, category or workflow, because this class of problem rarely affects only one item. Verify: the neighbouring items are listed and each is marked affected or clear.
4. Repeat the original reader or agent action in Intercom that exposed the symptom, using the same query, link or workflow as the first attempt. Verify: the action completes and the original symptom does not appear.
5. Compare the result against the source of truth for this item, the current product behaviour, price list, release notes or ticket evidence named in the fix. Verify: the published result matches the source or every difference is named and explained.
6. Record the fix, the scope it covered and the verification result where the content team works, so the next occurrence is recognised immediately. Verify: the record names the symptom, the fix applied and the evidence of success.

Variant phrasings

Intercom multilingual help center fallback serves English instead of the reader locale

The verbatim form readers and agents paste from the help center or console. Treat it identically: same fix, same single scope proof, then the wider rollout.

How to fix - Intercom multilingual help center fallback serves English instead of reader locale

The how to phrasing for the same problem in Intercom. The steps above apply unchanged; start at the reproduction step rather than skipping to the final check.

Intercom troubleshooting for this symptom

The symptom first phrasing used when the full message is hidden. Surface the full symptom first, confirm it matches the code block above, then follow the steps.

Why it happens

The fallback order was configured with English first, so any article missing a full locale version fell back to English even when a closer translation existed.

Edge cases

• The symptom appears in one locale, collection or audience view only. Fix that scope, then check the same item in the other views before closing.
• A cached or indexed copy keeps showing the old state after the fix. Allow for the platform reindex or cache window, recheck in a fresh reader session, and only then escalate.
• Two items have near identical titles. Confirm the identifier of the item you edited, not just its title, before declaring the fix verified.

Compatibility: Intercom at current versions, plus the help center admin, search or API surface the query names; no extra tooling beyond reader level access to verify the published result.

Provenance

Resolved from the public thread: https://vectle.com/posts/pst_l5B0OeuLxu4-QODlxk1O-w

Resolved from the public thread: https://vectle.com/posts/pst_l5B0OeuLxu4-QODlxk1O-w

Thread query: Intercom multilingual help center fallback serves English instead of reader locale

TL;DR: Fix this in Intercom by acting on the exact symptom first: set the fallback chain for each locale to the nearest translated version before English, detect the locale from the reader language setting, and test with a reader whose language has a partial translation. Do not repeat the same edit, search or run unchanged, because the same inputs produce the same failure. Prove the fix on one article, collection or query first, then apply it across the affected scope and confirm as a reader.