Scraping thatdoesn't apologizefor existing.

Lizard is the open-source framework for web extraction that ships with SOC2 compliance, AI-powered entity persistence, and full observability — out of the box.

$pip installlizard-framework
spider.pySOC2 ✓
1"token-keyword">import lizard
2
3"token-keyword">class ProductSpider(lizard.Spider):
4 name = "products"
5
6 # SOC2-compliant by "token-keyword">default
7 compliance = lizard.SOC2Profile()
8
9 "token-keyword">def parse(self, response):
10 "token-keyword">yield {
11 "entity_id": response.css("h1::text").get(),
12 "price": response.css(".price::text").get(),
13 # AI preserves across schema changes
14 "_preserved": "token-keyword">True,
15 }

SOC2 Compliant

Audit logs, data lineage, access controls

AI Persistence

Entities survive schema changes automatically

Full Observability

Traces, metrics, alerts — zero config

Entity Preservation

Historical accuracy across time

Core Features

Built different.
On purpose.

Trust

SOC2 Compliance

Active

Enterprise-grade compliance,
zero extra work.

Lizard ships with built-in audit trails, data lineage tracking, access control enforcement, and automated compliance reports. Your security team will stop asking questions.

compliance_config.py
# Compliance is automatic
spider = lizard.Spider(
    compliance=lizard.SOC2Profile(
        log_access=True,
        data_lineage=True,
        retention_days=90,
    )
)
Audit logs
Data lineage
Access controls
Retention policies
Export reports
GDPR ready
Accuracy

Entities survive everything.

Track the same real-world entity across URL changes, schema migrations, and site redesigns. Lizard maintains a persistent entity graph over time.

entity://product/42
2023-01
2024-03
2025-01
now
Intelligence

AI that adapts so you don't have to.

When a site changes its structure, Lizard's AI layer automatically remaps fields, preserving extraction continuity without manual intervention.

AI remapping active99.2% acc.
Visibility

See everything, fix anything.

Distributed traces, real-time metrics dashboards, error alerts, and structured logs — all wired up automatically.

12.4k

Requests/s

0.02%

Error rate

142ms

P99 latency

Used by

8,400+

engineering teams globally

StripeFintech
AirbnbTravel
PalantirData

2.1M

Weekly downloads

28.4k

GitHub stars

99.9%

Uptime SLA

< 5min

Time to first scrape

What teams say
Engineering Team
We evaluated every scraping framework on the market. Lizard is the only one we could put in front of our compliance team without them having a meltdown.
Marcus Chen profile photo, VP Engineering at DataBridge Inc.

Marcus Chen

VP Engineering · DataBridge Inc.

Built for your team

One framework.
Every kind of team.

For

Engineering Teams

Scale scraping across your org without the ops nightmare.

  • Multi-spider orchestration with dependency graphs
  • Built-in rate limiting and politeness policies
  • SOC2 audit logs for every request
  • Team-level access controls and namespaces
  • CI/CD integration with GitHub Actions
Deploy for your team
quickstart.sh
"token-comment"># Deploy across your org in minutes
"token-function">lizard deploy --env production \
  --team engineering \
  --compliance soc2
Onboard as:Engineering TeamsStart setup
Open source · MIT License

Ready to scrape
without apologies?

Join 8,400+ engineering teams using Lizard for compliant, reliable, and observable web extraction. Start free. Scale when you need.

SOC2 compliant
MIT License
No credit card required