Open Source · Apache 2.0
The slop filter your AI agents are missing
VectorLint is a programmable content review harness that turns your quality standards into measurable, source-grounded feedback for agents.
Platform · Coming Soon
The VectorLint platform is coming.
Run reviews from your terminal today. A hosted platform with dashboards, monitoring, and CI integration is on the way. Join the waitlist to be first to know.
We value your privacy and will never share your email with third parties.
Docs Quality You Can Measure
Shared Quality Standard
Define what good looks like as rules your agents and team can both check against.
Encode your style guides and quality standards as observable rules. VectorLint reviews content against them and returns grounded findings and scores, so humans and agents stay aligned on the same definition of quality.
Use "click" not "press"
Quality scores agents can optimize against.
Repeatable, comparable signals.
VectorLint turns review results into quality scores, computed from grounded findings, so agents can measure whether each revision actually improves content across review cycles.
Grounded feedback on every review
Every finding points back to the source.
VectorLint confirms each finding can be located in the content and drops the ones that can't, so agents get actionable, low-noise feedback instead of vague judgments to sift through.
I originally built VectorLint to help technical writers review content faster, because that was my biggest problem at the time. But as I spoke to more writers, I realized what they cared about wasn't speed. It was helping users with quality documentation, and proving that good docs drive business success.
VectorLint is where you define your quality standards and monitor your documentation health.
See what VectorLint finds in your docs
Install the CLIGive your agents feedback they can act on.
Also available as an open-source CLI.
FAQ