A WireMock simulation of Tavily's REST API — covering web search, content extraction, site crawling and mapping, and research tasks, plus credit usage, request logs, and feedback.
A shared instance of this template runs on a public endpoint — call it right now from curl, an SDK, or an AI agent. Responses are fixed examples and behavior can't be changed; rate limits apply.
Jump to the endpoints ↓Run this template as a mock API in WireMock Cloud for a private instance you control — edit stubs, add stateful scenarios, use dynamic response templating, and inspect requests in the log.
Open in WireMock Cloud (free) →Live sandbox · no signup, no API key
A running WireMock Cloud instance of this template, callable right now from a terminal, a script, or an AI agent.
Endpoints
POST Search for data based on a query [search]
curl -s -X POST 'https://78vyrd.wiremockapi.cloud/search' \
-H 'Content-Type: application/json' \
-d '{}'POST Retrieve raw web content from specified URLs [extract]
curl -s -X POST 'https://78vyrd.wiremockapi.cloud/extract' \
-H 'Content-Type: application/json' \
-d '{}'POST Initiate a web crawl from a base URL [crawl]
curl -s -X POST 'https://78vyrd.wiremockapi.cloud/crawl' \
-H 'Content-Type: application/json' \
-d '{}'Spec catalog — Tavily Search and Extract API (1.0.0) · 10 endpoints
| Method | Path | Summary |
|---|---|---|
| POST | /crawl | Initiate a web crawl from a base URL |
| POST | /extract | Retrieve raw web content from specified URLs |
| POST | /feedback | Submit feedback on a search request or session |
| POST | /logs | Get usage logs for your API keys |
| POST | /map | Initiate a web mapping from a base URL |
| POST | /org-usage | Get organization usage across all API keys |
| POST | /research | Initiate a research task |
| GET | /research/{request_id} | Get research task status and results |
| POST | /search | Search for data based on a query |
| GET | /usage | Get API key and account usage details |
Unauthenticated WireMock Cloud demo sandbox for Tavily — not an official Tavily sandbox, and it returns simulated example data only.
The template simulates api.tavily.com, the search API built for LLM agents. A search returns as many scored results as max_results asks for, adds an answer only when include_answer is set, and adds images, raw content and favicons only when their flags are. Each billable call is charged by Tavily's credit rules and added to the calling key's usage counters, so agent tools that watch their own budget can be tested too.
tavily-python accepts an api_base_url argument, so the client talks to the sandbox without any other change:
Agent frameworks that wrap the Tavily client, such as a LangChain or LlamaIndex search tool, pick up the same client configuration, so an agent can be pointed at the mock without code changes.
Fixed results keep agent evaluations comparable from run to run, and no credits are spent while a tool-calling loop is being built.
from tavily import TavilyClient
# api_base_url points the client at the sandbox;
# the key is never checked, but usage is counted per key
client = TavilyClient(
api_key="tvly-dev-any-value",
api_base_url="https://78vyrd.wiremockapi.cloud",
)
response = client.search("contract testing tools", max_results=3, include_answer=True)
print(response["answer"])
for result in response["results"]:
print(result["score"], result["url"])Use a URL on the reserved .invalid top-level domain, such as https://unreachable.invalid/page. It always lands in failed_results while the other URLs in the same request succeed.
Creating one answers 201 with status pending. The first poll answers 202 in_progress and the next one 200 completed with the report. An input containing [fail] produces a task that ends failed instead, and costs no credits.
No. Titles, URLs, snippets and reports are synthesized from your query. They have the right structure for parsing and citation code, not facts to reason about.
No. stream set to true on a research request is ignored and the normal JSON task comes back, since the spec documents no event-stream response for it.
Vendor names identify APIs represented by WireMock template sources. This page does not imply vendor endorsement, certification, partnership, or official integration status.