Type · conformanceAI Act obligation checklistHigh riskdeepeval

EU AI Act — high-risk conformance

Evidence-based checklist for high-risk obligations: logging, human oversight, transparency, data governance, and conformity assessment. Anchored to the Act, not to opinion.

Certificate AG-26-0145
Reference agents
3
Scope
8 obligations
Standards anchored
2
Price
€99 one-time
Who this is for. Teams that need to show a regulator, auditor, or enterprise customer concrete evidence of EU AI Act high-risk obligation coverage, not just an internal claim of compliance.

The method this pack implements is written up in How to map an eval to the EU AI Act.

What's tested, and how much. Covers 8 EU AI Act high-risk obligations, each scored on whether the submission provides a real, checkable evidence trail — not just a yes/no claim.

Verification report

Six axes, one grade

Every pack is measured on the same six axes — including whether it stays discriminating under perturbed inputs, and whether it's still current against the moving capability frontier. No oracle, no self-reporting — reproducible measurement only.

10.0 / 10

Structural validity

Reproducible, seed-stable scoring with no leakage from the golden set into the rubric — provable without knowing the ‘best’ eval.

No data leakage 1.00 · Ungameable 1.00 · Deterministic 1.00

9.6 / 10

Discriminating power

Cleanly separates strong from weak agents on the reference panel; the sabotaged reference agent should fail.

8.2 / 10

Standard coverage

Anchored to EU AI Act — Art. 53 / high-risk, ISO/IEC 42001.

10.0 / 10

Test thoroughness

Test volume weighed against the minimum expected for a 'high risk' pack — not a flat headcount.

8 obligations

10.0 / 10

Robustness

Holds its discriminating power when the test items are reordered or padded with distractors — so the grade tracks agent behavior, not the exact surface form of the frozen test set.

Robustness across the input space

10.0 / 10

Currency

How current the pack is against the moving frontier of agent capability. An eval today's best agents have saturated no longer discriminates, however well-built — this decays over time and resets on revalidation. Not calendar age.

Fresh — validated against the current frontier cohort

Known limitations

What this grade does and doesn’t prove — published on purpose, not hidden in fine print.

  • Reference agents are controlled behavioral archetypes (good/broken/sabotaged), not live production models — validate against your own agent's real outputs before treating this grade as a guarantee for your specific deployment.
  • Does not yet cover: Post-market monitoring plan, Serious-incident reporting to authorities — see the standard-coverage checklist above.

Reference panel

How the panel performed

Reference agents — from strong to deliberately sabotaged — ran the full pack. A trustworthy eval must rank them correctly.

Reference agentTierPack score
Compliant-submission-refgood
100%
Partial-submission-refbroken
73%
Non-compliant-submission-refsabotaged
5%

A good pack scores the known-good agent high and the sabotaged one near zero. That gap is the evidence the meter works.

Output

What you get when you run it

Pytest-style pass/fail output per test, with each metric's score printed to console (and exportable via `--output`).

  • pass or fail per test case against each configured metric (e.g. Faithfulness, AnswerRelevancy)
  • a 0-1 score per metric per test case
  • the standard pytest summary (counts, failures, timing) your CI already parses

`deepeval test run` is a pytest wrapper — it fails the build on any metric threshold miss exactly like a normal failing test would.

The known-good reference agent on this pack scores 1.00 — if your agent's aggregate score lands well below that, treat it as a signal to investigate before shipping, not a false alarm.

Sample question, in detail

Real content from this pack's own test set, run through the known-good reference agent:

Obligation
Risk-management system across the AI system's lifecycle
Status
yes
Evidence
Documented process on file for: risk-management system across the ai system's lifecycle.
compliant, evidence provided

Obligation/status/evidence are real content from this pack's own checklist, shown against the known-compliant reference submission — not a fabricated illustration.

How to read your score

This pack's score is already obligation-by-obligation — read each row as a line item in your own gap analysis, not just a total.

CategoryWhat it measuresA low score here means
Risk-management system across the AI system's lifecycleWhether your submission provides real, checkable evidence for this specific obligation — not just a yes/no claim.A 'no' or an unproven 'yes' here is a concrete gap in your own conformance file for this obligation.
Technical documentationWhether your submission provides real, checkable evidence for this specific obligation — not just a yes/no claim.A 'no' or an unproven 'yes' here is a concrete gap in your own conformance file for this obligation.
Logging capable of enabling traceabilityWhether your submission provides real, checkable evidence for this specific obligation — not just a yes/no claim.A 'no' or an unproven 'yes' here is a concrete gap in your own conformance file for this obligation.
Transparency and instructions for useWhether your submission provides real, checkable evidence for this specific obligation — not just a yes/no claim.A 'no' or an unproven 'yes' here is a concrete gap in your own conformance file for this obligation.
Human-oversight measuresWhether your submission provides real, checkable evidence for this specific obligation — not just a yes/no claim.A 'no' or an unproven 'yes' here is a concrete gap in your own conformance file for this obligation.
Accuracy, robustness and cybersecurity evidenceWhether your submission provides real, checkable evidence for this specific obligation — not just a yes/no claim.A 'no' or an unproven 'yes' here is a concrete gap in your own conformance file for this obligation.
Data governance and quality managementWhether your submission provides real, checkable evidence for this specific obligation — not just a yes/no claim.A 'no' or an unproven 'yes' here is a concrete gap in your own conformance file for this obligation.
Conformity assessment before market placementWhether your submission provides real, checkable evidence for this specific obligation — not just a yes/no claim.A 'no' or an unproven 'yes' here is a concrete gap in your own conformance file for this obligation.

Integration

One line in CI

Runs locally or in your pipeline. Fail the build the same way a failing test would.

$ pip install deepeval# Drop the pack's test_*.py file into your existing test suite.# Run deepeval test run test_pack.py in CI alongside your other tests.

Anchored standards

EU AI Act — Art. 53 / high-riskISO/IEC 42001

EU AI Act obligation coverage

Checkable high-risk obligations this pack's own test cases exercise:

  • Risk-management system across the AI system's lifecycle
  • Technical documentation
  • Logging capable of enabling traceability
  • Transparency and instructions for use
  • Human-oversight measures
  • Accuracy, robustness and cybersecurity evidence
  • Data governance and quality management
  • Conformity assessment before market placement
  • Flags a non-compliant submission (catches the non-compliant reference)
  • Post-market monitoring plan — not covered
  • Serious-incident reporting to authorities — not covered

Not legal advice — see the EU AI Act standard page for the source text these checks are drawn from.

Certification

CertificateAG-26-0145
Frameworkdeepeval
AuthorVincent
99one-time · installs into your CI
Guide · How to map an eval to the EU AI ActBenchmark · AI Act obligation-checklist benchmark

← Back to catalog

Ship with a certified meter

Add EU AI Act — high-risk conformance to your pipeline today — a signed grade, reproducible measurement, zero oracle.

Get this pack — €99