Google Custom Search API
Browse all Google APIs42
AI ReadinessFoundational (D-)42/100
Official vendor OpenAPI document · agent-readyData EnrichmentData TransformationapiKey2 EndpointsREST
For Agents
Run Google web searches scoped to a Programmable Search Engine and return ranked results with titles, snippets, and links so an agent can answer questions or feed retrieval pipelines.
Use for: I need to search the web for documentation about a product, Search our company sites for a specific term using a custom engine, Get the top 10 results for a query restricted to our docs domain, Find images matching a query using a custom search engine
Not supported: Does not crawl, index, or modify search results - use only to query a Programmable Search Engine and read ranked results.
Google Custom Search API (Programmable Search Engine) returns Google web search results scoped to one or more sites or to the wider web. It exposes a single endpoint that accepts a query, a custom search engine ID, and optional filters such as result language, safe-search level, country restriction, and start offset for pagination. A site-restricted variant returns only results from the configured engine's domains. Output includes title, snippet, link, and rich result metadata that match what end users would see on google.com.
Install Jentic One Beta
Connect the Custom Search API to your agent
Jentic One is a self-hosted execution layer for AI agents. It lets your agent call the Custom Search API, or any other public or private API you need. You set the rules, the agent never sees your credentials, and every call is logged.
Two steps, two machines. Install the instance in a safe environment, then register your agent from wherever it runs.
1
Step 1: Jentic One Host machine
# On the machine that will host your Jentic One instance:
curl -fsSL "https://jentic.com/install.sh?src=apis&api=%2Fapis%2Fgoogleapis.com%2Fcustomsearch" | sh2
Step 2: Agent machine
# On the machine where your agent runs (keep this separate from the instance):
curl -fsSL "https://jentic.com/install.sh?src=apis&api=%2Fapis%2Fgoogleapis.com%2Fcustomsearch" | sh
jentic register # connects your agent to your Jentic One instanceJentic One is in public beta. The setup above keeps your agent separate from the instance, which is what you want before using real credentials: an agent running as the same OS user as Jentic One can read its stored keys directly. Just evaluating? A single local install is fine to start. See the secure deployment guide for the tiers.
Capabilities
What an agent can do with Custom Search API.
Run a query against a custom search engine and return ranked web results
Restrict results to the sites configured on the engine using the siterestrict variant
Paginate through up to 100 results per query using start and num parameters
Filter by language, country, safe-search, and date range
Return image search results when the engine is configured for image search
Surface rich result metadata such as snippet, displayLink, and pagemap structured data
Use Cases
Patterns agents use Custom Search API for, with concrete tasks.
★ Site-Scoped Knowledge Lookup
A documentation chatbot uses Custom Search API with a Programmable Search Engine restricted to docs.example.com. The chatbot calls cse.siterestrict.list, takes the top three snippets, and grounds its response in the matching docs pages. Results return in well under a second per query.
Run a siterestrict search with cx='abc123' and q='refund policy', return the top 3 results' title, snippet, and link.
Web Retrieval for RAG
A retrieval-augmented generation pipeline calls the open Custom Search endpoint to fetch the latest web pages relevant to a user's question. The pipeline scrapes the returned links, embeds them, and supplies the chunks to a language model. Custom Search keeps freshness and ranking aligned with google.com results.
Search the web for 'gke 1.30 release notes', return the top 5 result links and snippets to feed into a RAG indexer.
Brand Monitoring Snapshot
A marketing team runs the Custom Search API daily for a list of brand and competitor terms, recording the top results' rank, link, and snippet in a warehouse. Trends in who appears for each term feed a brand-visibility dashboard. The 100-query free quota covers most small brand programmes.
For each term in ['acme cloud', 'acme storage'], retrieve the top 10 organic results and store rank, link, and snippet in BigQuery.
AI Agent Web Search Tool
An AI agent that needs current information about an external topic asks Jentic for the Custom Search API operation, supplies the cx and query, and receives results. Jentic isolates the API key in its vault, so the agent never holds the long-lived key directly.
Search the web for the current Kubernetes stable release using cx='abc123' and return the top 3 result snippets.
Key Endpoints
2 endpoints — google custom search api (programmable search engine) returns google web search results scoped to one or more sites or to the wider web.
METHOD
PATH
DESCRIPTION
GET
/customsearch/v1
Run a search query against a custom search engine
GET
/customsearch/v1/siterestrict
Run a search restricted to the engine's configured sites
GET
/customsearch/v1
Run a search query against a custom search engine
GET
/customsearch/v1/siterestrict
Run a search restricted to the engine's configured sites
Jentic AI Readiness Score
This API is usable in Jentic One now. Its AI-readiness score against Jentic's framework shows where it stands today and where improvements would make it even easier for agents to use.
C
Custom Search API
- Foundational (D-)42/100
87
Foundational Compliance56
Developer Experience & Jentic Compatibility19
AI-Readiness & Agent Experience94
Agent Usability100
Security38
AI DiscoverabilityPowered by JenticScoring Framework 1.0.0 | Scoring Engine 0.4.0
Show dimension breakdown
87
Foundational Compliance
Base layer of spec validity and structural soundness.
Grade: ASignals: 4
98%
Lint Results
Aggregated quality score from linter diagnostics, weighted by severity.
100%
Resolution Completeness
Percentage of `$ref` references that resolve successfully.
100%
Specification Validity
Checks whether the API description parses successfully and conforms to its declared specification (e.g., OpenAPI).
50%
Structural Integrity
Structural correctness score based on schema issues using logarithmic dampening.
56
Developer Experience & Jentic Compatibility
Clarity, completeness, and ingestion readiness for developers and tooling.
Grade: CSignals: 4
0%
Example Density
How richly the API is illustrated with examples.
100%
Example Validity
Percentage of examples that conform to their schemas.
25%
Response Coverage
Percentage of operations with complete response definitions (success, client error, server error).
100%
Tooling Readiness
Health of API ingestion, bundling, and resolution within Jentic pipelines.
19
AI-Readiness & Agent Experience
Semantic breadth, depth, and agent comprehension for AI systems.
Grade: FSignals: 4
76%
Description Coverage
Coverage of descriptions across API elements.
0%
Error Standardization
Coverage of RFC 9457 Problem Details for error responses.
0%
OperationId Quality
Coverage, uniqueness, and casing consistency of operationIds for AI inference.
0%
Summary Coverage
Coverage of summaries across operations/tags/info.
94
Agent Usability
Functional utility, complexity comfort, and AI orchestration readiness.
Grade: A+Signals: 1
94%
Complexity Comfort
Agent comfort level based on API operational and structural complexity.
100
Security
Trust, risk posture, and security compliance.
Grade: A+Signals: 1
100%
Authentication Strength
Average quality of security schemes based on authentication method strength (weakest link for OAuth2).
38
AI Discoverability
Findability, semantic richness, and reasoning readiness.
Grade: FSignals: 1
38%
Descriptive Richness
Clarity and depth of descriptions across API elements.
Score it yourself
Every API in the directory is allowlisted, so you can re-score it with no key required.
npx @jentic/api-scorecard-cli score <openapi-url>Why Jentic?
What agents get from Jentic-routed access to this vendor.
Setup
Wiring the Custom Search API by hand means obtaining an API key, keeping it out of source, and appending it to the 'key' query parameter on every request to customsearch.googleapis.com along with your Programmable Search Engine cx. Through Jentic you install once, import the Custom Search API from the API Directory, store the API key once, and your agent calls it.
Permission scoping
This API reads ranked results and does not expose write or delete operations, so scoping is about the operations you allow: limit the agent to the operations it needs, such as running a web search or a site-restricted search. Both are read-only queries against your configured search engine.
Credential isolation
Your Custom Search API key is stored once, encrypted, by your own Jentic One instance and injected into the key query parameter at execution time. It never enters the agent's prompt, logs, or context.
Intent-based discovery
Agents search Jentic by intent such as 'run a Google web search', and Jentic returns the matching Custom Search operation with its input schema, including cx, q, num, start, and safe, so the agent calls the right endpoint without browsing the reference docs.
Related APIs
Alternatives and complements available in the Jentic catalogue.
FAQs
Specific to using Custom Search API through Jentic.
What authentication does the Custom Search API use?
The Custom Search API uses an API key passed as the 'key' query parameter. Through Jentic the API key is stored encrypted in your Jentic One instance and injected into requests at execution time, so the raw key never appears in the agent's context.
Can I get more than 10 results per query from the Custom Search API?
Yes, in pages. Each call returns at most 10 results; set the 'start' parameter (1, 11, 21...) and 'num' (max 10) to paginate. The API supports retrieving the first 100 results for a given query.
What are the rate limits for the Custom Search API?
The Custom Search API allows 100 free queries per day per project, with paid tiers available for higher volume up to 10,000 queries per day per engine. Per-second QPS limits also apply; check the Programmable Search Engine pricing page for current ceilings.
How do I run a Google web search through Jentic?
Run pip install jentic, search Jentic for 'google custom search query', load the schema for the cse.list operation on customsearch.googleapis.com, and execute it with your cx (engine ID) and q (query) parameters.
Does the Custom Search API let me search the entire web like google.com?
Only when the linked Programmable Search Engine is configured to search the whole web. By default an engine targets a specific list of sites; toggle 'Search the entire web' in the Programmable Search Engine console to widen the scope.
Can I limit what my agent is allowed to do with the Google Custom Search API?
Yes. Because you run Jentic One yourself, your own rules decide which operations and credentials the agent may use. The Custom Search API is read-only with just two operations, so you can restrict the agent to only what it needs: a general web search or the site-restricted search against your configured Programmable Search Engine. You allow both, one, or neither, and the agent can call nothing else since there are no write or delete operations to expose.
GET STARTED