---
title: "DeepEval - Air Market"
description: "Changing an agent's prompt or tools is guesswork unless you can measure it. This sets up evaluation with DeepEval: building test cases, running them, seeing what failed, and using those failures to d…"
canonical: "https://airmarket.ai/app/deepeval"
---

[Air Market](/) / [Data & Analytics](/category/data) / DeepEval

Skill by [Confident AI](/@confident-ai)

# DeepEval

Turn eval failures into the next change you make

## What this does

Changing an agent's prompt or tools is guesswork unless you can measure it. This sets up evaluation with DeepEval: building test cases, running them, seeing what failed, and using those failures to decide what to fix next. Reach for it when you cannot tell whether your last change helped.

### Top 3 things it does for you

- Add evals to an agent or LLM app
- Generate a test dataset to evaluate against
- Run an eval suite and see what failed

## Information

**DeepEval at a glance**

|  |  |
| --- | --- |
| Kind | Skill |
| Publisher | [Confident AI](/@confident-ai) |
| Category | [Data & Analytics](/category/data) |
| Works with | Agent Skills - an open standard supported by 40+ agents |
| Version | 1.0.0 |
| Size | 58 KB |
| Licence | Apache-2.0 |
| Repository | [confident-ai/deepeval](https://github.com/confident-ai/deepeval/tree/571c1e0cac97580444fbb06db721539c2863aa84/skills/deepeval) |
| GitHub stars | 17,817 (as of Nov 2026) |
| Installs | 2,913, counted by skills.sh (Nov 2026) |

## Security report

**Verified.** AIR scanned this exact version of DeepEval and flagged nothing. AIR's verdict: **safe**.

**What DeepEval's check was run against**

|  |  |
| --- | --- |
| Commit | 571c1e0cac97580444fbb06db721539c2863aa84 |
| Content hash | c30c8cd5af7d0681e526d19b776f25c15b01ee9f2f6fb39a540493d316492615 |
| Files read | [Browse the files at this exact commit](https://github.com/confident-ai/deepeval/blob/571c1e0cac97580444fbb06db721539c2863aa84/skills/deepeval/SKILL.md) |

A scan is named by the exact content it read, not by a version number. A different version is a different artifact, and nothing here applies to a version AIR has not scanned.

[Read the full security report for DeepEval](/security/deepeval)

## About DeepEval

This is the DeepEval workflow for evaluating AI agents and LLM applications - agents being the main case, alongside tool-using workflows, multi-turn chatbots, and RAG pipelines.

It covers the loop end to end: generating datasets and goldens, writing pytest eval suites with the Python SDK, running them from the CLI, traced evals, and reporting results to Confident AI for monitoring, online evals, and trace inspection.

The point is the iteration that follows - taking eval failures back into prompts, tools, retrieval, and agent behaviour. Instrumenting an app with DeepEval tracing belongs to the deepeval-tracing skill, and raw OpenTelemetry export without the package to deepeval-otel.

## What's inside

- SKILL.md - instructions
- LICENSE - reference
- references/artifact-contracts.md - reference
- references/choose-use-case.md - reference
- references/confident-ai.md - reference
- references/datasets.md - reference
- references/intake.md - reference
- references/iteration-loop.md - reference
- references/metrics.md - reference
- references/pytest-e2e-evals.md - reference
- references/synthetic-data.md - reference
- references/traced-evals.md - reference
- templates/metrics.py - script
- templates/test_multi_turn_e2e.py - script
- templates/test_single_turn_no_tracing.py - script
- templates/test_single_turn_tracing.py - script

## Questions

### Is DeepEval safe to install?

AIR scanned this exact version of DeepEval and flagged nothing. AIR's verdict: safe. The verdict is named by the content that was read, at commit 571c1e0cac97, so it cannot drift onto a version nobody scanned. The full report is at https://airmarket.ai/security/deepeval.

### How do I install DeepEval?

Open https://airmarket.ai/app/deepeval on Air Market and use the install action: Air Market hands Claude a prompt naming this exact skill, and Claude sets it up. Source, pinned to the exact commit this listing describes: https://github.com/confident-ai/deepeval/tree/571c1e0cac97580444fbb06db721539c2863aa84/skills/deepeval. Install from that link, not the branch.

## Related on Air Market

- [More Data & Analytics listings](/category/data)
- [Everything by Confident AI](/@confident-ai)

---

Canonical URL: <https://airmarket.ai/app/deepeval>