How is Autocomplete API usage measured?
Successful normalized operations consume the units defined by the active plan. Request headers make usage and rate-limit state visible to your backend.
Free tier · no payment method required
Autocomplete API
Get marketplace-specific suggestions for a partial phrase. Call one documented GET endpoint across 13 Amazon marketplaces with the same API key, normalized response envelope, and production metadata.
TL;DR
The Autocomplete API returns structured JSON under data.amazonSearchAutocompleteResults. It is available through REST, GraphQL, and MCP, with a 1 hour cache policy and explicit request, freshness, rate-limit, and usage metadata.
Use this Glade operation through
Live demo
Try the Autocomplete API in your browser
Run the allowlisted autocomplete preset against the real Glade backend. Parameters are predefined and validated before an upstream request is made.
Try the Autocomplete API
Run the predefined request before writing a single line
example response · ready
{
"data": {
"amazonSearchAutocompleteResults": [
{
"suggestion": "mechanical keyboard"
},
{
"suggestion": "mechanical keyboard wireless"
},
{
"suggestion": "mechanical keyboard switches"
}
]
}
}Endpoint
What does the Autocomplete API return?
The Autocomplete operation is available through every Glade interface and resolves through the same validation and normalized output contract.
REST responseavailable
GET /api/amazon/autocomplete returns data.amazonSearchAutocompleteResults in the standard response envelope.
GraphQL fieldavailable
amazonSearchAutocompleteResults exposes the same normalized operation through the typed GraphQL schema.
MCP toolavailable
get_amazon_autocomplete exposes the allowlisted operation to compatible AI agents.
Developer First
How do you call the Autocomplete API in seconds?
Use /api/amazon/autocomplete without maintaining marketplace-specific acquisition, parsers, retry rules, or response adapters.
One schema, every operation
Use this API alongside 16 other Glade operations without changing authentication or the response envelope.
Metadata, not guesswork
Every response includes request, cache, freshness, rate-limit, and usage metadata for production decisions.
See the contract before coding
Copy a request in Python, TypeScript, or cURL and inspect the example JSON before moving into production.
example response · ready
import requests
response = requests.get(
'https://gladeapi.com/api/amazon/autocomplete',
params={'searchTerm': 'mechanical key', 'domain': 'US'},
headers={'API-KEY': 'glade_live_YOUR_API_KEY'}
)
data = response.json()[ .JSON ]
{
"data": {
"amazonSearchAutocompleteResults": [
{
"suggestion": "mechanical keyboard"
},
{
"suggestion": "mechanical keyboard wireless"
},
{
"suggestion": "mechanical keyboard switches"
}
]
}
}One API, 13 marketplaces
How does Glade normalize autocomplete across Amazon marketplaces?
Amazon storefronts vary by host, currency, locale, delivery context, and available fields. Glade keeps those differences behind one versioned autocomplete contract.
One schema, every marketplace
One documented contract.
13 Amazon storefronts use one marketplace code, validation layer, and output contract.
data.amazonSearchAutocompleteResults
marketplace: US
success: true
Currency and locale included
Context stays attached.
Marketplace, currency, delivery context, request ID, cache state, and fetch time travel with the normalized response.
cache: HIT
fetched_at: 2026-08-16T12:00:00Z
usage_units: 1
REST, GraphQL, and MCP
One key, every interface.
Keep one server-side API key while application code and agents use the interface that fits their workflow.
REST: /api/amazon/autocomplete
GraphQL: amazonSearchAutocompleteResults
MCP: get_amazon_autocomplete
Operation-level freshness
Caching fits the data.
This endpoint uses a 1 hour cache policy so volatility and request cost stay explicit.
cache_ttl: 3600
rate_limit: documented
request_id: glade_...
Glade Amazon API17 endpoints supported
View docs GET/api/amazon/autocomplete?searchTerm=mechanical+key&domain=US
searchTermmechanical keydomainUS
curl 'https://gladeapi.com/api/amazon/autocomplete?searchTerm=mechanical+key&domain=US' \
-H 'API-KEY: glade_live_YOUR_API_KEY'example response · ready
{
"data": {
"amazonSearchAutocompleteResults": [
{
"suggestion": "mechanical keyboard"
},
{
"suggestion": "mechanical keyboard wireless"
},
{
"suggestion": "mechanical keyboard switches"
}
]
}
}Compare
How does Glade's Autocomplete API compare to a self-managed integration?
Glade packages autocomplete acquisition, validation, normalization, caching, errors, and request metadata behind one server-side key.
FeatureGlade APISelf managed
AuthenticationOne API-KEY headerSource-specific credentials or sessions
Setup requiredCreate a key and call the endpointBuild acquisition, parsing, retries, and adapters
Marketplace support13 stores through one contractSeparate host and locale handling
Response schemadata.amazonSearchAutocompleteResultsSource-specific page structures
Freshness1 hour policy with metadataYour team owns caching and observation time
InterfacesREST, GraphQL, MCP, and SkillYour team maintains each adapter
Can I use autocomplete data in production?
Use public data, respect source rules and third-party rights, keep keys server-side, and match retention and downstream processing to your approved use case.
FAQ
Have a question? We got answers
Answers about this Glade API, its parameters, interfaces, freshness, and production use.
How do I call the Autocomplete API?
Create a Glade key, send a GET request to /api/amazon/autocomplete with one of the documented identifier or query presets, and read data.amazonSearchAutocompleteResults from the response envelope.
What does the Autocomplete API return?
Get marketplace-specific suggestions for a partial phrase. Optional fields remain optional, and every response includes request and freshness metadata needed for production handling.
Which marketplaces support autocomplete?
Use the same operation across 13 supported Amazon marketplaces by passing the documented domain code.
How fresh is autocomplete data?
This operation uses a 1 hour cache policy. Read the cache-state and data-fetched-at headers instead of inferring freshness from when your application received the response.
Can I use the Autocomplete API with an AI agent?
Yes. The get_amazon_autocomplete MCP tool and generated documentation expose the same allowlisted operation and validation rules.
Can I use the Autocomplete API in production?
Yes. Keep keys server-side, validate optional fields, store request IDs, use bounded retries for temporary failures, and review source rules and third-party rights for your use case.
Contract statistics
API contract at a glance
Current Autocomplete details generated from Glade's versioned operation catalog.
13
marketplaces
1 hour
cache policy
3+
interfaces
Ready to call the Autocomplete API?
Create a key, keep it server-side, and run the predefined request in minutes.
Get an API key