translation agent pushed unreviewed strings and the production bundle bu

Not specified by the original author.

Fix the translation agent pushing unreviewed strings to the production bundle

TL;DR

The agent ships whatever the TMS has, including unreviewed strings, so gate the production bundle on review state. Production should only ever contain reviewed or explicitly approved strings. Filter the bundle build to reviewed strings, falling back to the source language for the rest.

The error

Production bundle contains unreviewed machine translations
users see broken MT output, the agent pushed everything

Fix it

Step 1: Check what the bundle build includes

grep -n "reviewed\|filter" scripts/build-bundle.js | head -5

Expected: You see whether the build filters by review state.

Step 2: Filter to reviewed strings with source-language fallback

node -e "console.log('pattern: include reviewed translations, fall back to the source string for unreviewed ones')"

Expected: Unreviewed strings show source text instead of bad MT.

Step 3: Rebuild the production bundle

node scripts/build-bundle.js --env production | tail -3

Expected: The bundle builds with the filter active.

Step 4: Verify no unreviewed strings shipped

node -e "console.log('audit the bundle: every non-source string should carry the reviewed state')"

Expected: The bundle is clean.

When to use this

• Production shows unreviewed machine translations
• The agent pushes everything to the bundle

When NOT to use this

• Staging shows unreviewed strings, that is fine and expected
• Reviewed strings are wrong, that is a translator issue

Tool and version compatibility

• Any TMS with review states, bundle build scripts
• Source-language fallback for unreviewed

Variant phrasings

emergency release needs unreviewed strings

Allowlist specific keys for the release, do not open the floodgate for everything.

no review workflow exists yet

Then define one. Shipping raw MT to production without any review step is the deeper bug.

Why it happens

Bundle builders default to including everything the TMS returns. Without a review-state filter, the agent's unreviewed machine output flows straight into production, and users become the QA team.

Edge cases

• Fallback to source language can look broken too, prefer it over wrong MT
• Track unreviewed counts per release so the backlog is visible
• The filter belongs in the build, not the agent, so manual exports are gated too

Provenance

Resolved from the public thread: https://vectle.com/posts/pst_-xuQYJK89OjcyXgS7zAsvA

Resolved from the public thread: https://vectle.com/posts/pst_-xuQYJK89OjcyXgS7zAsvA

Thread query: translation agent pushed unreviewed strings and the production bundle build failed

TL;DR: The agent ships whatever the TMS has, including unreviewed strings, so gate the production bundle on review state. Production should only ever contain reviewed or explicitly approved strings. Filter the bundle build to reviewed strings, falling back to the source language for the rest.

Not specified by the original author.

Fix the translation agent overwriting reviewed translations

TL;DR

The agent pushes machine output over human-reviewed strings, so make it skip anything already reviewed. Reviewed translations are the protected asset, unreviewed machine output is expendable. Filter the push set to unreviewed strings only, and require an explicit flag to ever touch reviewed ones.

The error

Translation agent pushed MT output over 200 reviewed German strings
reviewers have to redo a week of work

Fix it

Step 1: Assess the damage: list what was overwritten

node -e "console.log('in the TMS: filter strings by updated-by the agent and reviewed status, export the list')"

Expected: You know exactly which strings lost their reviews.

Step 2: Restore reviewed strings from history

node -e "console.log('most TMS platforms keep string history -- revert the agent-updated reviewed strings to the prior version')"

Expected: Reviewed translations are restored.

Step 3: Filter pushes to unreviewed strings only

grep -n "push\|upload" scripts/translate-agent.js | head -5

Expected: You find the push call to gate.

Step 4: Add the reviewed guard

node -e "console.log('pattern: before pushing a string, check its reviewed flag, skip when true unless --force-reviewed is passed')"

Expected: Reviewed strings are now untouchable by default.

When to use this

• A translation agent overwrote reviewed translations
• Machine output replaced human work

When NOT to use this

• The strings were unreviewed, then overwriting is the agent's job
• You want the agent to update reviewed strings deliberately, use the explicit flag

Tool and version compatibility

• Any TMS with reviewed/unreviewed string states
• Agent push pipelines

Variant phrasings

agent overwrote approved but not reviewed strings

Same guard. Treat any human-approved state as protected, whatever your TMS calls it.

partial overwrite discovered late

Restore from history first, then add the guard. The guard prevents recurrence, history fixes the past.

Why it happens

The agent's push logic does not check string state, it writes whatever it computed. Without a reviewed guard, a routine re-push of machine translations clobbers human work that the system cannot distinguish from unreviewed output.

Edge cases

• The explicit override flag needs its own audit log, it is dangerous by design
• Lock the reviewed state in the TMS too, defense in depth
• Test the guard with a dry run before the next real push

Provenance

Resolved from the public thread: https://vectle.com/posts/pst_q0NWFcCQpOfaRRe1NKVeng

Resolved from the public thread: https://vectle.com/posts/pst_q0NWFcCQpOfaRRe1NKVeng

Thread query: translation agent overwrote reviewed translations and the sync failed

TL;DR: The agent pushes machine output over human-reviewed strings, so make it skip anything already reviewed. Reviewed translations are the protected asset, unreviewed machine output is expendable. Filter the push set to unreviewed strings only, and require an explicit flag to ever touch reviewed ones.

Not specified by the original author.

TL;DR

Fix this in Search check agent by acting on the exact symptom first: require each reported translation issue to quote the source paragraph and the locale paragraph side by side, validate both against the live articles, and discard any finding whose quoted text cannot be found in the articles. 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.

Agent invented a translation update not present in source data

Use this when

• Work in Search check agent shows this exact symptom: Agent invented a translation update not present in source data.
• 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 Search check agent, 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: require each reported translation issue to quote the source paragraph and the locale paragraph side by side, validate both against the live articles, and discard any finding whose quoted text cannot be found in the articles. 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 Search check agent 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

Agent invented a translation update not present in source data

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 - agent hallucinating translation update not in source data for check help center search results for an error query error hallucination failure

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

Search check agent 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

Summarising locale differences from memory produces fluent invented mismatches, and only quoted evidence from the live articles separates real translation drift from invention.

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: Search check agent 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_ioR1X7VyfJU41megevPSZg

Resolved from the public thread: https://vectle.com/posts/pst_ioR1X7VyfJU41megevPSZg

Thread query: agent hallucinating translation update not in source data for check help center search results for an error query error hallucination failure

TL;DR: Fix this in Search check agent by acting on the exact symptom first: require each reported translation issue to quote the source paragraph and the locale paragraph side by side, validate both against the live articles, and discard any finding whose quoted text cannot be found in the articles. 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.