Agent Evals
Evaluation framework for grading computer-use agent performance using captured traces.
What it does
Evaluation framework for grading computer-use agent performance using captured traces.
- searchable work memory
- meeting notes
- CRM updates
- SOP generation
- workflow automation
- task mining
- agent evaluation
- computer-use agents
- Category
- AI evaluation and observability
- Users
- individuals, small teams, enterprises
- Public API
- Not offered
Not yet disclosed: deployment options, platforms, languages and support channels.
Pricing
- Free plan$0
- Subscription$21/month
Free trial on paid plans
Prices as published by screenpipe, checked Oct 2026.
Security and compliance
As reported by screenpipe. Ask for audit reports before you buy.
- SOC 2 Type IICompliant
- Local-first by default — data stays on-device
- Encrypted at rest with zero-knowledge key hierarchy
- PII redaction via on-device AI model
- Opt-in capture controls with per-app and per-URL filtering
- Open source / source-available for independent audit
- Confidential-inference PII removal in trusted execution environment (TEE)
Reviews
No reviews yet
Reviews come from people who use Agent Evals at work. Share how your team uses it to help other buyers decide.
Alternatives
Tokenshire
by Olam Labs
Simulated town of AI agents running autonomously for long horizons, socializing and managing government, economy, and resources under constraints.
KitchenBench
by Robocurve
Evaluating robots on ten bimanual kitchen tasks, including pouring, lid removal, folding, part-mating, two-arm handover, and tool-mediated scooping.
LLM evaluation
by Openlayer
Rigorous evaluation for GenAI and LLMs—test every prompt, trace every response, and validate every output.
Clarity
by Guide Labs
The first inherently interpretable AI platform, including PromptConceptTraining to understand which part of the prompt is responsible for the output.
Web-based AI simulation platform to build, evaluate, and test AI agents against realistic scenarios and custom metrics.
Loss Analyses
by AfterQuery
Identifies where and why models fail in professional contexts through systematic study, pinpointing the precise failure modes and distributional gaps that inform dataset and env...
Is this your product?
Claim it to keep details accurate and submit evidence for the next ranking review.
Claim Agent Evals