Google Safe Browsing API
Browse all Google APIs37
AI ReadinessNon-Ready (F)37/100
Official vendor OpenAPI document · agent-readySecurityThreat DetectionapiKey1 EndpointsREST
For Agents
Check whether URLs are flagged by Google Safe Browsing using the v5 hash-prefix lookup endpoint, with privacy-preserving partial-hash lookups.
Use for: Check whether a user-submitted URL is flagged by Safe Browsing, Look up a hash prefix against Google's threat list, Find the threat types associated with a specific URL hash, Validate links inside an inbound email body before delivering it
Not supported: Does not block traffic, render warning interstitials, or scan attachments - use for hash-prefix URL reputation lookups against Google threat lists only.
The Google Safe Browsing API v5 lets client applications check whether URLs and other web resources match Google-curated lists of malware, phishing, and other unsafe content. The v5 surface focuses on a hash-prefix search endpoint that preserves user privacy by sending only a hash prefix to the server. Safe Browsing is licensed for non-commercial use; Web Risk is the commercial equivalent for products that sell or generate revenue from the threat data.
Install Jentic One Beta
Connect the Safe Browsing API to your agent
Jentic One is a self-hosted execution layer for AI agents. It lets your agent call the Safe Browsing 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%2Fsafebrowsing" | 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%2Fsafebrowsing" | 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 Safe Browsing API.
Check a URL hash prefix against Google's malware and phishing threat lists
Identify Safe Browsing threat types associated with returned full hashes
Operate the lookup with API key authentication suitable for backend services
Use partial hash lookups to avoid sending full URLs to the server
Validate URLs in user-generated content before publishing or following them
Build crawler-side allow and deny logic against the Safe Browsing threat list
Use Cases
Patterns agents use Safe Browsing API for, with concrete tasks.
★ Pre-delivery URL scanning for messaging platforms
A messaging service hashes every URL inside an inbound message and checks the prefix against Safe Browsing v5 before delivery. If the hash returns a matching threat, the service quarantines the message and notifies the sender. Hash-prefix lookups keep the full URL out of Google's view, supporting privacy-sensitive deployments.
Compute SHA-256 over each URL, take the 4-byte prefix, GET /v5/hashes:search with the prefix, and quarantine messages whose URLs return a matching threat.
User-generated content moderation
A content platform feeds every newly posted URL into Safe Browsing and removes posts whose URLs match malware or phishing lists. The hash-prefix model lets the platform check millions of URLs daily without revealing the underlying content to Google, and the v5 lookup is fast enough for inline use during post submission.
On each new post, compute the hash prefix for embedded URLs and call GET /v5/hashes:search; remove the post if any threat type is returned.
Crawler safety gate
A web crawler checks every fetch target against Safe Browsing before issuing the request, preventing the crawler from following links into known malware. The lookup happens once per host and is cached, so the crawl rate stays high while exposure to drive-by downloads is reduced.
Before each fetch, GET /v5/hashes:search with the URL hash prefix; skip the URL if the response includes any threatType.
Agent-driven URL safety checks through Jentic
An AI agent built on Jentic exposes URL safety as a tool: when asked to summarise a webpage, it first calls Safe Browsing through Jentic, refuses to fetch flagged URLs, and explains why. Jentic stores the API key in the vault and returns the matching operation by intent.
Search Jentic for 'check url with safe browsing', execute GET /v5/hashes:search with the URL hash prefix, refuse to summarise the page if any threat is returned.
Key Endpoints
1 endpoints — the google safe browsing api v5 lets client applications check whether urls and other web resources match google-curated lists of malware, phishing, and other unsafe content.
METHOD
PATH
DESCRIPTION
GET
/v5/hashes:search
Search hash prefixes against Safe Browsing threat lists
GET
/v5/hashes:search
Search hash prefixes against Safe Browsing threat lists
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.
S
Safe Browsing API
- Non-Ready (F)37/100
100
Foundational Compliance56
Developer Experience & Jentic Compatibility14
AI-Readiness & Agent Experience94
Agent Usability100
Security50
AI DiscoverabilityPowered by JenticScoring Framework 1.0.0 | Scoring Engine 0.4.0
Show dimension breakdown
100
Foundational Compliance
Base layer of spec validity and structural soundness.
Grade: A+Signals: 4
99%
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).
100%
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.
14
AI-Readiness & Agent Experience
Semantic breadth, depth, and agent comprehension for AI systems.
Grade: FSignals: 4
57%
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).
50
AI Discoverability
Findability, semantic richness, and reasoning readiness.
Grade: C-Signals: 1
50%
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 Safe Browsing API by hand means provisioning a Google API key, appending it to every request, and formatting hash-prefix lookups against the single v5 endpoint on safebrowsing.googleapis.com. Through Jentic you install once, import the Safe Browsing API from the API Directory, store the API key once, and your agent calls it.
Permission scoping
This API exposes a single read-only lookup operation (GET /v5/hashes:search), so a rule limits the agent to that one operation: it can check URL reputation against Google's threat lists and nothing else. There are no state-changing operations to add, so the scope stays read-only by construction.
Credential isolation
Your Safe Browsing API key is stored once, encrypted, by your own Jentic One instance and appended to the request at execution time. The key never enters the agent's prompt, logs, or context.
Intent-based discovery
Agents search Jentic by intent such as 'check a URL with Safe Browsing', and Jentic returns the GET /v5/hashes:search operation with its input schema so the agent calls the right endpoint without reading the developer guide.
Related APIs
Alternatives and complements available in the Jentic catalogue.
FAQs
Specific to using Safe Browsing API through Jentic.
What authentication does the Safe Browsing API use?
The v5 spec exposes a single GET /v5/hashes:search endpoint accessed with a Google API key passed via the standard key query parameter. Through Jentic, the API key is stored encrypted in the vault and injected into requests at execution time, so the agent never holds the raw key.
Can I check a URL for phishing with the Safe Browsing API?
Yes. Compute SHA-256 over the canonicalized URL, send a hash prefix on GET /v5/hashes:search, and inspect the response for matching full hashes and their threat types. SOCIAL_ENGINEERING in the threat type indicates phishing; MALWARE indicates a malicious binary or script.
What are the rate limits for the Safe Browsing API?
Google publishes a default quota of 10,000 queries per day per project for Safe Browsing v5, with a per-second cap of approximately 10 requests per second. Higher quotas are not generally granted because Safe Browsing is restricted to non-commercial use; commercial deployments should use the Web Risk API instead.
How do I check a URL through Jentic?
Search Jentic for 'check url with safe browsing', load the schema for GET /v5/hashes:search, compute the SHA-256 hash prefix client-side, and execute the call. Jentic returns the parsed response so the agent can inspect threatType values for each match.
Is the Safe Browsing API free?
Yes, but only for non-commercial use as defined in Google's terms. If you sell or generate revenue from the threat data, use Google's Web Risk API instead, which is the paid commercial equivalent and is also covered in this catalogue.
Why does Safe Browsing only send hash prefixes to the server?
The hash-prefix model preserves user privacy: the client sends a 4-byte prefix of the SHA-256 hash of the URL rather than the URL itself. The server returns matching full hashes for that prefix, and the client compares locally to determine whether the original URL is on the list - so Google never sees the full URLs being checked.
Can I limit what my agent is allowed to do with the Safe Browsing API?
Yes. Because you run Jentic One yourself, your own rules decide which operations and credentials the agent may use. The Safe Browsing API exposes a single read-only lookup, GET /v5/hashes:search, so you can scope the agent to that one operation and nothing else. It can check URL hash prefixes against Google's malware and phishing threat lists, and since there are no state-changing operations to enable, the access stays read-only.
GET STARTED