ToolkitAPI — Python SDK

Official Python SDK for ToolkitAPI.io — a family of 16 focused HTTP APIs covering DNS, email, images, PDFs, scraping, screenshots, AI-powered text tools, barcodes, finance data, media extraction, webhooks, and more.

Installation

pip install toolkitapi

Requires Python 3.9+ and httpx.

Quick start

Use a single toolkit directly:

from toolkitapi import DNS

dns = DNS(api_key="tk_...")
print(dns.lookup(domain="example.com"))
dns.close()

Or the unified client to access multiple toolkits with a single API key:

from toolkitapi import ToolkitAPI

with ToolkitAPI(api_key="tk_...") as tk:
    dns_result = tk.dns.lookup(domain="example.com")
    scores = tk.textanalysis.readability(text="The quick brown fox.")
    png_bytes = tk.screenshot.capture(url="https://example.com")

Toolkit instances are created lazily on first property access and reused thereafter. close() (or leaving the with block) tears down every HTTP client that was opened.

Toolkits

Toolkit	Class	Highlights
AI-Text	AiText	LLM-powered extraction, classification, sentiment, summarization
Auth	Auth	JWT create/decode, TOTP, OAuth helpers, API-key generation
Barcode	Barcode	Barcode and QR code generation, decoding, raw-image downloads
Convert	Convert	Format (JSON/YAML/CSV/XML) and unit conversions
DevTools	DevTools	JSON/YAML/XML validators, regex tester, UUID, hashing, fake data
DNS	DNS	DNS records, WHOIS, propagation, email auth, typosquat, SSL certs
Email	Email	Validation, deliverability, disposable detection, SPF/DMARC
Finance	Finance	Currency conversion, FX rates, VAT lookup
Geo	Geo	IP geolocation, reverse geocoding, distance/bearing, timezone
Image	Image	Resize, convert, optimise, analyse, remove background
Media	Media	YouTube metadata/transcripts, universal media extraction, crawling
PDF	PDF	Generate, merge, split, extract, stamp, protect
Scrape	Scrape	Web scraping, readability, meta, SEO audits, broken-link checks
Screenshot	Screenshot	Page/element/HTML/template rendering to PNG/JPG/PDF (sync + async)
TextAnalysis	TextAnalysis	Readability, PII masking, profanity, similarity, summarize, language
Webhook	Webhook	Request bins, HTTP mocks, replay captured requests

Full endpoint reference at https://toolkitapi.io/docs.

Scrape toolkit examples

The scrape SDK now mirrors the current unified scrape API. You can fetch page content, request structured extraction, render JavaScript pages, and run specialized sitemap, robots, crawl, and PDF helpers from one client.

Unified page scrape

from toolkitapi import Scrape

scrape = Scrape(api_key="tk_...")

result = scrape.fetch(
    url="https://example.com/blog/post",
    output="markdown",
    extract={
        "article": True,
        "links": True,
        "meta_tags": True,
    },
)

print(result["content"])
print(result.get("article"))
scrape.close()

JavaScript rendering

with Scrape(api_key="tk_...") as scrape:
    result = scrape.render_page(
        url="https://example.com/app",
        wait_until="networkidle",
        output="text",
        scroll=True,
    )
    print(result["content"])

PDF, sitemap, and robots helpers

with Scrape(api_key="tk_...") as scrape:
    pdf = scrape.pdf_extract(url="https://example.com/report.pdf", pages="1-2")
    sitemap = scrape.parse_sitemap("https://example.com/sitemap.xml", limit=100)
    robots = scrape.parse_robots_txt("https://example.com")

AI extraction and chunking

with Scrape(api_key="tk_...") as scrape:
    result = scrape.ai_extract(
        url="https://example.com/product",
        schema={
            "type": "object",
            "properties": {
                "name": {"type": "string"},
                "price": {"type": "string"},
            },
        },
        prompt="Extract the product name and displayed price.",
        render_js=True,
    )
    print(result.get("ai_extract"))

Authentication

Every request is authenticated by the X-API-Key header. Provide your key when constructing a toolkit (or the unified ToolkitAPI client); it is reused for every request in that session.

tk = ToolkitAPI(api_key="tk_live_...")

Error handling

All non-2xx responses raise an exception that inherits from toolkitapi.ToolkitAPIError:

Exception	HTTP status
AuthenticationError	401, 403
ValidationError	400, 422
NotFoundError	404
RateLimitError	429
ServerError	5xx
APIError (fallback)	anything else — exposes .status_code and .detail

from toolkitapi import DNS, NotFoundError, RateLimitError

dns = DNS(api_key="tk_...")
try:
    dns.lookup(domain="does-not-exist.example")
except NotFoundError:
    ...
except RateLimitError:
    ...

Async jobs (Screenshot, Media)

Some endpoints can enqueue work and return a job descriptor instead of bytes. Pass async_mode=True to the relevant method, then poll the returned job ID:

job = tk.screenshot.capture(url="https://example.com", async_mode=True)
result = tk.screenshot.wait_for_job(job["job_id"], timeout=60)
png_bytes = tk.screenshot.download(result["object_name"])

Binary downloads

Endpoints that return raw image/PDF bytes expose an optional output_path= kwarg — pass a filename to stream the response directly to disk instead of loading it into memory:

tk.barcode.qr_raw_image(data="https://toolkitapi.io", output_path="qr.png")

Configuration

• timeout — per-client request timeout (seconds). Defaults to 30s; 60s for heavier toolkits (Image, PDF, Scrape, Screenshot).
• base_url — override the API host (useful for staging or self-hosted deployments).

dns = DNS(api_key="tk_...", timeout=10.0, base_url="https://staging.dns.toolkitapi.io")

Development

cd sdk/python
pip install -e ".[test]"
pytest

License

MIT — see LICENSE in the repository root.