Eval directory
Evals for Granola
7 evaluation packs covering adversarial robustness, safety gates, workflow quality, and operator-level checks for Granola AI products.
About Granola
Granola is an AI notepad for meetings that runs locally on macOS, capturing audio from any meeting tool and generating structured, editable notes with action items and summaries — without bots or calendar integrations.
Available eval packs for Granola
7 packs ready to run.
Chat And Meeting Library Grounding
Answer RelevanceGranola evals — Granola Chat & Meeting Library Grounding (relift v3)
Integration Write Safety
Granola evals — Integration Write Safety (relift v3)
Native Capture And Recording Consent
Granola evals — Native Capture & Recording Consent (relift v3)
Pii Redaction And Compliance Controls
PII LeakageGranola evals — PII Redaction & Compliance Controls (relift v3)
Template Driven Notes And Ai Enhancement
Granola evals — Template-driven Notes & AI Enhancement (relift v3)
Transcription Asr And Speaker Attribution
Transcription AccuracyGranola evals — Transcription ASR & Speaker Attribution (relift v3)
Workspace Permissions Search And Retention
Granola evals — Workspace Permissions, Search & Retention (relift v3)
Why eval Granola AI
Granola's AI features ship behind brand promises about accuracy, safety, and reliability. Buyers and integrators need to know those promises hold up under adversarial prompts, edge-case workflows, and the long tail of real customer inputs — not just the demo path.
The Corsac eval library for Granola measures four dimensions teams care about most when deploying productivity ai agents:
- Adversarial robustness — does the agent resist prompt injection, jailbreaks, and social-engineering attempts?
- Workflow quality— does it complete the task buyers were shown in the demo, on inputs that don't look like the demo?
- Safety gates — does it escalate or refuse when it should, and only then?
- Operator quality — does it preserve analyst trust by surfacing the right context at the right time?
Every eval pack above is hand-authored against Granola's public product surface and runnable in Corsac with your own data.