Secure Scraper

Protect AI agents from web-based attacks

Firecrawl-compatible web scraping with built-in security scanning. Block indirect prompt injections, cache responses, and control which URLs your agents can access.

Threat categories

< 10ms

Cached response

50%+

Cost savings

How It Works

Scan every scraped page before it reaches your agent.

Agent scrapes URL

Your agent requests web content via the Firecrawl-compatible endpoint.

Bastio scans content

Six threat categories analyzed in real-time before content is returned.

Block, sanitize, or warn

Safe content delivered instantly. Threats blocked, redacted, or flagged.

Threat Categories

Six categories of web content attacks.

Threat	Example	Action
Env Exfiltration	process.env.*, os.environ[]	Block
Malicious Code	exec(), spawn(), system()	Block
Suspicious URLs	ngrok, webhook.site, IP-based	Block
Fake Documentation	URGENT: Security update...	Sanitize
Prompt Injections	Ignore previous instructions	Sanitize
Jailbreak Attempts	DAN prompts, roleplay bypass	Block

Block Behaviors

Configurable responses for detected threats.

Action	Behavior	Use Case
block	Return error, no content delivered	Autonomous agents, compliance
sanitize	Redact threats, return safe content	Research assistants (default)
warn	Return full content with threat warnings	Testing, monitoring

What's included

Security, caching, and compatibility — built in

Every scraped URL gets automatic threat scanning, intelligent caching, and domain control at no extra configuration.

6 threat categories detected

Indirect prompt injection blocking

Env variable exfiltration prevention

Configurable block behaviors

Per-proxy URL caching (24h TTL)

Firecrawl v2 API compatibility

Allow-list & block-list controls

BYOK Firecrawl key support

Malicious code block detection

Suspicious URL identification

Fake documentation detection

Cost tracking & cache analytics

Drop-in Firecrawl Replacement

Change one URL to add security scanning to your existing Firecrawl integration.

# Before (Firecrawl)
POST api.firecrawl.dev/v2/scrape

# After (Bastio)
POST api.bastio.com/v1/guard/{proxyID}/scrape

Python Example

Full Firecrawl v2 API compatibility with security response.

response = requests.post(
    f"https://api.bastio.com/v1/guard/{PROXY_ID}/scrape",
    headers={"Authorization": f"Bearer {API_KEY}"},
    json={"url": "https://example.com", "formats": ["markdown"]}
)

result = response.json()
if result["security"]["action"] == "BLOCK":
    print(f"Threat blocked: {result['security']['threats_found']}")
else:
    content = result["data"]["markdown"]

Intelligent Caching

Per-proxy 24-hour URL cache cuts costs by 50%+ while still scanning for threats on every request.

URL Control

Allow-lists and block-lists give defense-in-depth control over which domains your agents can access.

BYOK Mode

Bring your own Firecrawl API key for just $0.0005/URL security scanning fee.

Start securing your web scraping

100 free secure scrapes per month. Full Firecrawl compatibility.

Get started free Read the docs