Libb evaluates text, not individuals.
The score speaks to the design of the communication, not the ability of the reader. A low score is information about the message, not the people receiving it.
Linguistic Ergonomics · Patent-pending
Can a reader actually process your message?
Most writing tools help you get a message correct or simple. Libb evaluates whether a reader can process it, and shows you the one change that makes it land. It is a clinically derived estimate of processing burden, not a guarantee of comprehension.
Built by a licensed speech-language pathologist. A person stays in the loop on every message.
What Libb measures
The Cognitive Processability Score.
Grade-level tools count syllables and sentence length. The Cognitive Processability Score evaluates the cognitive operations a reader must perform to extract meaning, and produces a clinically derived record of how the communication was evaluated.
CPS is not a readability score.
It evaluates how clauses connect to ideas, where ambiguity forces a reread, and where the message asks the reader to infer specifics that were never named. It correlates only weakly with grade-level scoring.
Source content is not stored.
Submitted text is evaluated and discarded. The report ID, score, and pattern record are retained on the issuing account for personal trending.
CPS is the regulatory record.
The score is a clinically derived evaluation record that complements current AI regulations covering AI-generated workplace communication. Libb does not provide legal counsel.
01 · Origin
A speech-language pathologist's answer to AI-generated communication.
Built by a clinician, for the people everyone else designs around.
Samantha Cocroft is a licensed Speech-Language Pathologist with more than a decade of clinical practice alongside neurodivergent populations, from children to working professionals: ADHD, autism, and executive function differences. Her patients spent years teaching her what accessible communication actually looks like, and what happens when it does not arrive.
The pattern in her practice was consistent. A piece of writing that other people scanned in under a minute took her patients an hour. Three follow-up messages to clarify what the first one was asking. A stress response that lasted the rest of the day. None of this was a personal deficit, and none of it was unusual. It was a design failure, repeated.
When AI writing tools began producing professional communication at a volume no individual editor could review, Samantha saw the same clinical pattern appearing in workplace email, performance reviews, policy memos, and notice letters. The scale changed. The pattern did not. Libb applies the cognitive-communicative assessment framework she spent a decade refining, called Linguistic Ergonomics, and scores whether AI-generated communication actually fits the brain it was sent to.
Responsible AI
Built for the human in the loop.
Most systems are deployed without anyone measuring what they cost the people who have to read, understand, and act on the output. These are the commitments Libb is willing to be held to. We lead with what we can prove.
In the loop
Human in the loop by design
A person decides and sends every message. Libb shows information about a communication; it does not make decisions about people, and it does not send anything for you.
Clinical
Clinically grounded
A licensed speech-language pathologist sets the reference the scoring is calibrated and checked against. Human clinical judgment is the ground truth; the system applies it at scale rather than replacing it.
Governed
Binding commitments
Our ethics are written into our governing documents. Our Ethical Charter sets hard limits, including that outputs may not be used to make adverse decisions about people.
Standards
Aligned to recognized frameworks
We align with the NIST AI Risk Management Framework and are working toward ISO/IEC 42001, with HITRUST as a complementary path. We claim only what we have earned.
Privacy
Privacy by design
We evaluate the message, not the person. Content is processed transiently, scored, and discarded. We do not use submitted content to train our models.
Limits
What we will not claim
We do not claim the evaluation is perfect, publish accuracy figures we cannot substantiate, or display certifications we have not earned. A person is always in the loop.
Pricing · Early access
Three ways to work with Libb.
A self-serve plan for the professional reviewing AI-generated drafts every day. A five-day Communication Assessment for organizations that need a clinically derived evaluation record. An enterprise relationship for platforms generating communication at scale.
Practitioner
$99/month
For the HR partner, manager, or communications professional reviewing AI-generated drafts every day.
- Unlimited CPS evaluations
- Side-by-side rewrite suggestions
- Personal score history and PDF export
- Single-seat access
Most popular
Communication Assessment
$5,000/assessment
The founder personally evaluates real workplace communication against the Linguistic Ergonomics framework. The premium reflects clinical review, not product access.
- Up to 50 AI-generated messages evaluated by the founder
- Founder-signed clinical sign-off on the findings
- Tailored recommendations for your communication corpus
- Evaluation record with regulatory context
- Delivered in five business days
Enterprise & API
Let’s talk
For platforms generating communication at scale, and organizations that need continuous evaluation rather than point-in-time.
- API access with workspace controls
- Multi-tenant configuration
- Clinically derived evaluation records
- Custom band thresholds and reporting
| Capability | Practitioner | Communication Assessment | Enterprise & API |
|---|---|---|---|
| CPS evaluation | Unlimited, self-serve | Up to 50 messages | Unlimited, programmatic |
| Pattern reports | Per message | Aggregate plus per message | Per message and per tenant |
| PDF scorecard | Self-export | Assessment-prepared | API on demand |
| Evaluation record | Personal record | Yes, clinician-signed | Yes, custom |
| Founder review | Not included | Yes, included | Yes, on engagement |
| Custom band thresholds | Default bands | Default bands | Yes, multi-tenant |
| Methodology versioning | Latest | Pinned at delivery | Customer-pinned |
Contact us
Tell us about the communication you are working on.
Conversations welcomed with HR teams, communications leaders, and platform teams working on internal communication, regulatory readiness, or the cognitive accessibility of AI-generated content. Fill in a few details and your email app opens with a message to the founder, ready to send.
