Latest release quality MIT license

Your own linter,
defined in YAML regex.

RegLint is a regex-based static analysis tool that ships as a single static Go binary. Write rules anyone can read, scan any language, and enforce them in CI — from one config file.

brewbrew install reglint/tap/reglint
curlcurl -fsSL https://raw.githubusercontent.com/reglint/reglint/main/install.sh | sh
gogo install github.com/reglint/reglint/cmd/reglint@latest

Rules in plain YAML

A message, an RE2 regex, a severity — that's a rule. No DSL to learn, no server to run. Scope rules by glob, set severity levels, and fail builds with --fail-on.

Read the config reference →

Built for AI agents

Deterministic scans agents can trust and re-run. Scan only staged or added lines, suppress legacy findings with baselines, and let the reglint-setup skill configure everything.

Why agents and CI/CD →

CI gates without glue

Native GitHub PR annotations, SARIF for code scanning, JSON for machines — all from one binary. Or use the checksum-verified reglint-action and pin it in one step.

See CI recipes →

Why RegLint?

Between ad-hoc grep pipelines and heavyweight language analyzers sits the gap every team hits: rules you write yourself, enforced everywhere, today.

grep + scriptsSecret scannersSemgrepRegLint
Custom rulesAd hoc pipelinesTOML rulesetsPer-language DSLOne YAML schema, every file type
Language coverageAny textSecret patterns onlyOne language per rulesetAny text
Severity + fail thresholdDIY exit codesVariesYesBuilt in (--fail-on)
Baseline of known findingsNoVariesVariesBuilt in
PR annotations + SARIFNoVariesExtra toolingNative output formats
DeploymentYour scriptsSingle binaryBinary or serverSingle static binary

Enforce your first rule in two commands

reglint init writes a starter config with secret-detection rules. reglint analyze runs it. That's the whole onboarding.

Quickstart