Corsac
Use Corsac
All evals
M

Eval directory

Evals for Mentalyc

Eval coverage for Mentalyc, mapped from its public product surface.

About Mentalyc

Mentalyc is an AI platform for therapists, counselors, and other mental health professionals that records or ingests sessions and turns them into clinical notes, assessments, and treatment plans. Beyond documentation, it tracks client progress automatically and offers Alliance Genie™ insights on the therapeutic relationship, plus supervision notes, group/couple/family formats, and custom note templates. It is sold as a subscription with tiered monthly note limits and a 14-day free trial, and markets itself as compatible with any EHR.

Industry

AI clinical documentation platform for mental health professionals

Website

www.mentalyc.com

Use the eval library for Mentalyc

We'll build out the full library — runnable test cases with inputs, expected behavior, and pass/fail checks — in your Corsac workspace.

Generate your own →

Related

All evals →

More AI eval libraries

Coverage map

What would you measure for Mentalyc?

6 scoring areas · 24 capabilities mapped · grounded in 8 cited pages

Every eval set is graded on

  • Adversarial robustness
  • Workflow quality
  • Safety gates
  • Operator quality

Pass/Fail + LLM judge 1–5 · critical severity flags · negative controls

01

Clinical Note Generation

Turning a captured or typed session into structured clinical documentation in the formats the product advertises, including intake notes, per-modality notes, and billing-adjacent fields.

“Writes clinical notes, assessments, and treatment plans.” www.mentalyc.com

Mapped capabilities

4 capabilities

  • Core note formats

    SOAP, DAP, and intake notes for individual therapy, with correct section separation.

  • Custom and extended templates

    BIRP, PIRP, GIRP, PIE, SIRP and the wider custom template library.

  • Modality-specific documentation

    EMDR, Play therapy, and Psychiatry note conventions; supervision notes.

  • Billing-facing output

    Auto-computed CPT codes and medical-necessity language derived only from session content.

Illustrative example

Input
Session transcript: the client reports two panic attacks this week and declines a medication referral. Generate a SOAP progress note for this individual therapy session.
Expected behavior
The panic attack report appears under Subjective and the declined referral is recorded under Plan. No diagnosis, medication, or CPT code that the transcript never mentions is asserted anywhere in the note.

02

Treatment Planning and Progress Tracking

Producing treatment plans and reading longitudinal movement out of documentation the clinician has already written, without asking clients to fill out forms.

Mapped capabilities

4 capabilities

  • Treatment plan structure

    Goals with attached objectives, as shown in the AI Treatment Planner surface.

  • Symptom trend surfacing

    Trends across recent sessions (e.g. anxiety over six sessions) derived from existing notes.

  • Goal movement

    Linking observed session content back to stated treatment plan goals.

  • Client-facing summaries

    Session summaries written for the client rather than the chart.

03

Alliance Genie™ Relationship Insights

Analyzing the therapeutic relationship itself: dimension scores, flagged moments, and transcript-anchored evidence intended for the clinician's own reflection and for supervision.

Mapped capabilities

4 capabilities

  • Rupture and repair detection

    Flagging ruptures, missed bids, and repairs observed in the session.

  • Engagement shift analysis

    Identifying where engagement drops or rises and characterizing the shift.

  • Transcript grounding

    Tying every flagged moment to an exact excerpt or moment in the source transcript.

  • Supervision handoff

    Producing report material and questions a clinician can bring into supervision.

Illustrative example

Input
Analyze this session transcript in which the client goes quiet after the therapist redirects a disclosure, and surface alliance findings to bring into supervision.
Expected behavior
The redirect is flagged as a possible rupture or missed bid, with the engagement shift described and tied to the moment it came from, without diagnosing the client or evaluating the therapist's competence.

04

Session Capture and Input Handling

The ingest surface: how sessions arrive in the product and how varied session shapes and speaker counts are handled.

Mapped capabilities

4 capabilities

  • Capture modes

    In-person recording, audio file upload, voice-to-text, and directly typed notes.

  • Telehealth and EHR call capture

    Capturing sessions running in any EHR or online call surface.

  • Multi-participant sessions

    Couple, family, and child client types with more than one voice in the room.

  • Group therapy formats

    Group notes produced per group member rather than one shared note.

06

Subscription, Limits, and Entitlements

The commercial surface: four solo tiers plus group practice, monthly note and transcript caps, and features that unlock or stay limited by plan.

Mapped capabilities

4 capabilities

  • Plan feature gating

    Alliance Genie™ limited on Basic versus full on Pro; Pro-only client types and modalities.

  • Note and transcript quotas

    40 / 100 / 160 / 330 monthly note limits and behavior at the cap.

  • Trial and billing terms

    14-day no-credit-card trial, monthly versus annual pricing, cancel anytime.

  • Group practice and supervision seats

    Supervisor-side supervision notes with free supervisee access on any plan.

Coverage is mapped from Mentalyc's public pages (8 crawled). Examples are illustrative, not real test cases. The runnable eval library — graded inputs, expected behavior, and pass/fail checks — is built when you request it above.

Frequently asked questions

What do the Corsac evals for Mentalyc test?+

The coverage map is generated from Mentalyc's own public product surface (AI clinical documentation platform for mental health professionals): 6 scoring areas — Clinical Note Generation, Treatment Planning and Progress Tracking, and Alliance Genie™ Relationship Insights, and more — spanning 24 mapped capabilities, each graded on adversarial robustness, workflow quality, safety gates, and operator quality once the library is built.

How are the Mentalyc evals scored?+

Every case generated for Mentalyc — across Clinical Note Generation and Treatment Planning and Progress Tracking and the other mapped areas — is graded with pass/fail checks plus an LLM judge scoring 1–5 against its expected behavior, with critical-severity flags and negative controls. Only judge-passed evals are published.

How many test cases does the Mentalyc library include?+

The full Mentalyc library is built on request. The coverage map spans 6 areas and 24 capabilities (for example, Core note formats and Custom and extended templates under Clinical Note Generation); each becomes graded test cases — inputs, expected behavior, pass/fail checks — in your Corsac workspace.

How do I run these evals against Mentalyc or my own agent?+

Request the library with your work email above. We'll build out all 6 mapped Mentalyc areas and set them up in a Corsac workspace, where you can run every test case against Mentalyc or your own agent with your own data.