Overview
LexLint is a compliance lint for AI, scraping, privacy, cybersecurity, communications, content-moderation, product-liability, age-gating, and reuse law, and all nine topics are matched by the lint rather than held for reference. A coding agent building an app that crawls the web, generates content, or deploys a chatbot connects to the LexLint MCP server, declares what the app does and where it will operate, and gets back lint-style findings: which obligations apply in each jurisdiction, with citations and freshness dates.
The framing matters: LexLint is a lint, not a legal assessment. Like a code linter, it catches basic issues early in the development cycle. It never certifies compliance, and it replaces neither review by qualified counsel nor QA. Its passing state is "no basic issues found," never "compliant."
Findings come from the LexLint law library. AI and scraping law are tracked across every country, US states, and the EU, maintained continuously. Privacy, cybersecurity, communications, content-moderation, product-liability, age-gating and reuse law are researched a jurisdiction at a time rather than tracked everywhere, expanding through the same research pipeline. Every entry carries its citation, an as-of date, and a staleness flag, and a jurisdiction LexLint has no current data for is reported as a warning, never silently passed.
Available Tools
→ check_access Key, quota and law library freshness, checked first
→ claim_trial_key A 30-day trial key with no account, when the developer picks that route
→ set_profile Declare what your app does and where it runs
→ get_law AI, scraping, privacy, cybersecurity, communications, content-moderation, product-liability, age-gating, and reuse law for one place
→ run_lint Activities × jurisdictions → lint findings
→ resolve_domain_jurisdiction Domain → whose law governs its operator
→ submit_feedback Send feedback on LexLint to the people who build it
→ upload_lint_run Store this run on the portal, only when you say yes
How It Works
Agent → LexLint: run_lint(
activities=["crawls_web", "generates_content"],
jurisdictions=["us", "eu", "kr"]
)
# LexLint returns findings, not verdicts
WARN eu TDM opt-outs are enforceable rights reservations (DSM Art. 4(3))
WARN eu GPAI training-data policy obligations may apply (AI Act Art. 53)
INFO kr AI Framework Act: generated-content labeling duties phase in
0 errors, 2 warnings, 1 info: review the warnings below
# Every finding carries a citation, an as-of date, and a staleness flag
Tech Stack
LexLint runs as a standalone edge worker with no server behind it, and reads the law library exclusively through UnGovr's public keyed API, the same one any developer can use. Bring an UnGovr API key; LexLint passes it through and never stores it.
Links
Product site: https://lexlint.io
MCP endpoint: https://mcp.lexlint.io/mcp
API keys: https://www.ungovr.org/open-data/api-keys