Labsco
brave logo

spellcheck

✓ Official★ 159

by brave · part of brave/brave-search-skills

USE FOR spell correction. Returns corrected query if misspelled. Most search endpoints have spellcheck built-in; use this only for pre-search query cleanup or "Did you mean?" UI.

🔥🔥✓ VerifiedFreeQuick setup
#git#search
🧩 One of 7 skills in the brave/brave-search-skills package — works on its own, and pairs well with its siblings.

This is the playbook your agent receives when the skill activates — you don't need to read it to use the skill, but it's here to audit before installing.

Spellcheck

Requires API Key: Get one at https://api.search.brave.com

Plan: Included in the Spellcheck plan. See https://api-dashboard.search.brave.com/app/subscriptions/subscribe

Endpoint

GET https://api.search.brave.com/res/v1/spellcheck/search

Authentication: X-Subscription-Token: <API_KEY> header

Parameters

ParameterTypeRequiredDefaultDescription
qstringYes—Query to spell check (1-400 chars, max 50 words)
langstringNoenLanguage preference (2+ char language code, e.g. en, fr, de, pt-br, zh-hans). 51 codes supported
countrystringNoUSSearch country (2-letter country code or ALL)

Response Fields

FieldTypeDescription
typestringAlways "spellcheck"
query.originalstringThe input query as submitted
resultsarraySpell-corrected suggestions. May be empty when no correction is found
results[].querystringA corrected version of the query

Example Response

{
  "type": "spellcheck",
  "query": {
    "original": "artifical inteligence"
  },
  "results": [
    {
      "query": "artificial intelligence"
    }
  ]
}

Use Cases

  • Pre-search query cleanup: Check spelling before deciding which search endpoint to call
  • "Did you mean?" UI: Show users a corrected suggestion before running the search
  • Batch query normalization: Clean up user inputs in bulk

Notes

  • Built-in alternative: Web Search and LLM Context have spellcheck=true by default — use this standalone endpoint only when you need the correction before searching
  • Context-aware: Corrections consider the full query context, not just individual words