Assess your software for legal risks

An assessment is LexLint's scanner at work. Your coding agent declares what the app does and where it operates, and LexLint reads that declaration against the law library and returns the legal risks it finds: the obligations that apply, each with its citation. It is free, and it runs from Claude Code, Codex or any other agent that speaks MCP.

What goes in, what comes out

In: a declaration

The activities the app performs, the jurisdictions it operates in and the roles it plays under the law. That is the whole input: no source code, schema or data leaves the repository.

read against the law library

Out: findings

One for each obligation that applies: the law, what it asks, where, its citation and the date the library last read the source. The same declaration against the same library returns the same findings.

One assessment

declared
generates_content, crawls_web
operates
EU, Canada, US-California
law library
WARNEU AI Act, Article 50since Providers of AI systems that interact directly with people must ensure users are informed they are dealing with an AI system unless obvious from context, and providers of generative AI must mark synthetic audio, image, video, or text output in a machine-readable, detectable format.read against its source
WARNCanada Uber Technologies Inc. v. Hellersince The joint reasons held that unconscionability requires both an inequality of bargaining power and a resulting improvident bargain.read against its source
WARNUS-California SB 942since A covered provider must make a disclosure verification tool available at no cost to the user, one that lets a user assess whether image, video, or audio content was created or altered, other than by a minor modification, by the provider's generative AI system and that outputs any system provenance data detected in the content.read against its source , for counsel

3 findings: 2 for code, 1 for counsel

Paste the first line into the agent you already have open. See what a whole run returns

Where it lives

DevPack runs the assessment in your coding CLI today. Its documentation is on the developer site:

Keeping the declaration true: LexCode

An assessment reads what the declaration says. LexCode reads the code, on your machine, and in CI it flags code that does something the declaration does not say, with the file and the line. It is free and open source. What it does, a short video and the languages it reads

An assessment catches the basics early. Its findings are legal information with citations, not legal review.