> ## Documentation Index
> Fetch the complete documentation index at: https://docs.agentserp.com/llms.txt
> Use this file to discover all available pages before exploring further.
# Search
Search for information with various retrieval strategies.
## Example Request
```bash theme={null}
curl -X POST https://api.agentserp.com/search \
-H "Authorization: Bearer " \
-H "Content-Type: application/json" \
-d '{"query": "openai"}'
```
## OpenAPI
````yaml POST /search
openapi: 3.0.0
info:
title: AgentSerp API
version: 1.0.0
description: API for search, extraction, summarization, and more.
servers:
- url: https://906e-2001-5a8-6cc-e000-55c0-f0b-9da7-10b0.ngrok-free.app
description: AgentSerp API
security: []
paths:
/search:
post:
summary: Search across the web
requestBody:
content:
application/json:
schema:
type: object
properties:
query:
type: string
description: Natural-language search query
mode:
type: string
enum:
- auto
- keyword
- neural
default: auto
description: >-
Retrieval strategy: let backend choose, force keyword, or
force embedding
depth:
type: string
enum:
- basic
- advanced
default: basic
description: Depth/quality tier; advanced costs more & may run longer
topic:
type: string
enum:
- general
- news
default: general
description: Domain-specific retrieval agent
max_results:
type: integer
minimum: 1
maximum: 50
default: 10
description: Maximum number of results to return
add_answer:
anyOf:
- type: boolean
enum:
- false
- type: string
enum:
- basic
- type: string
enum:
- detailed
default: false
description: >-
`false` → no answer. `basic` → one-sentence LLM answer.
`detailed` → richer
include_raw_content:
type: boolean
default: false
description: If true, include cleaned HTML text in each result
include_images:
type: boolean
default: false
description: If true, also run an image search
recency:
anyOf:
- type: string
enum:
- day
- week
- month
- year
description: Human shortcut for common recency windows
- type: integer
minimum: 0
exclusiveMinimum: true
description: Limit results to this many days in the past or a preset
filters:
type: object
properties:
include_domains:
type: array
items:
type: string
description: Restrict results to these hostnames
exclude_domains:
type: array
items:
type: string
description: Omit any result whose hostname matches
include_text:
type: array
items:
type: string
description: Require these tokens appear in document body
exclude_text:
type: array
items:
type: string
description: Reject docs containing these tokens
description: Fine-grained filtering for /search
view:
type: array
items:
type: string
enum:
- results
- answer
- images
- embeddings
description: Selective blocks to include in response
default:
- results
description: Which data blocks to include in the response
similar_to:
type: string
format: uri
description: Return docs similar to this URL instead of keyword search
stream:
type: boolean
description: If true, respond with NDJSON streaming
response_format:
type: string
enum:
- json
- ndjson
- csv
- parquet
description: Serialization format for large/bulk results
user_context:
type: object
additionalProperties:
nullable: true
description: Opaque object echoed for personalization
required:
- query
description: Input contract for POST /search
responses:
'200':
description: Search results
content:
application/json:
schema:
type: object
properties:
request_id:
type: string
description: Echo of client-supplied ID or server-assigned UUID
data:
type: object
properties:
answer:
type: string
description: LLM answer if requested
results:
type: array
items:
type: object
properties:
title:
type: string
description: Human-readable title of the document
url:
type: string
description: Canonical URL of the document
content:
type: string
description: Snippet or full text, depending on request
score:
type: number
minimum: 0
maximum: 1
description: Relevance score (0-1) where 1 is most relevant
meta:
type: object
additionalProperties:
nullable: true
description: >-
Backend-specific key-value pairs (author,
publishedDate, etc.)
required:
- title
- url
description: Generic doc object used across endpoints
description: Ordered list of search hits
images:
type: array
items:
type: object
properties:
url:
type: string
format: uri
alt:
type: string
required:
- url
description: Image URL plus alt text
embeddings:
type: array
items:
type: array
items:
type: number
description: Vector embeddings in the same order as `results`
description: Endpoint-specific payload
meta:
type: object
properties:
api_version:
type: string
description: >-
Semantic version string of the API that produced the
response
processing_ms:
type: integer
description: Time taken on the server to satisfy the request, in ms
tokens_used:
type: integer
description: Total LLM tokens consumed (prompt+completion)
cost:
type: object
properties:
total:
type: number
minimum: 0
description: Total dollar cost for serving this request
breakdown:
type: object
additionalProperties:
type: number
minimum: 0
description: >-
Optional map of operation → incremental cost (e.g.
neuralSearch: 0.004)
required:
- total
description: Per-request cost information
required:
- api_version
description: Metadata block attached to every response
required:
- data
description: Universal response wrapper
'400':
description: Bad request
content:
application/json:
schema:
type: object
properties:
error:
type: string
required:
- error
````