eu-ai-act-mcp

Assess Art. 6(3) 'No Significant Risk' Exception

euaiact_assess_art6_3_exception

Read-onlyIdempotent

Determine if an Annex III AI system is exempt from high-risk classification under Art. 6(3) by validating the four exception conditions and profiling exclusion.

Instructions

Walk through the Art. 6(3) exception for Annex III high-risk AI systems. An Annex III system is NOT high-risk only if the provider affirmatively assesses no significant risk to health, safety, or fundamental rights, AND the system falls under one of the four conditions: (a) narrow procedural task, (b) improves prior human activity, (c) detects patterns without replacing human assessment, (d) preparatory task. Set no_significant_risk_to_health_safety_fundamental_rights=true only when that threshold has been assessed. CRITICAL: The exception does NOT apply if the system performs profiling of natural persons (Art. 6(3), third subparagraph). Providers invoking the exception must document the assessment (Art. 6(4)) and still register in the EU database (Art. 49(2)).

Input Schema

TableJSON Schema

Name	Required	Description
`annex_iii_number`	No	Annex III category number (1-8). Provide if known - helps the response reference the right rules.
`preparatory_task`	No	Art. 6(3)(d): intended to perform a preparatory task for an assessment relevant to an Annex III use case.
`performs_profiling`	Yes	Does the system perform profiling of natural persons? If true, Art. 6(3) exception is UNAVAILABLE regardless of other conditions (Art. 6(3), third subparagraph).
`documented_assessment`	No	Has the provider documented the Art. 6(3) exception assessment? Required by Art. 6(4) regardless of which condition applies.
`narrow_procedural_task`	No	Art. 6(3)(a): intended to perform a narrow procedural task (e.g. transform unstructured to structured data, classify documents, detect duplicates).
`improves_prior_human_activity`	No	Art. 6(3)(b): intended to improve the result of a previously completed human activity (e.g. improve drafting style).
`detects_patterns_without_replacing_human_review`	No	Art. 6(3)(c): intended to detect decision-making patterns or deviations, not replace/influence human assessment without proper review.
`no_significant_risk_to_health_safety_fundamental_rights`	No	Has the provider affirmatively assessed that the system does not pose a significant risk of harm to health, safety, or fundamental rights, including by not materially influencing decision-making outcomes? Required for Art. 6(3).

Output Schema

TableJSON Schema

Name	Required	Description	Default
`reasoning`	Yes
`registration_duty`	Yes
`relevant_articles`	Yes
`exception_available`	Yes
`conditions_evaluated`	Yes
`documentation_reminder`	Yes
`profiling_blocks_exception`	Yes

Schema Changelog

Changes observed during successful MCP inspections.

Changed2 schema fields changedv1.5.0Aug 19, 2026

changedInput schema / properties / annex_iii_number / description

Previous value: -"Annex III category number (1-8). Provide if known — helps the response reference the right rules."New value: +"Annex III category number (1-8). Provide if known - helps the response reference the right rules."

changedInput schema / properties / performs_profiling / description

Previous value: -"Does the system perform profiling of natural persons? If true, Art. 6(3) exception is UNAVAILABLE regardless of other conditions (Art. 6(3) second subparagraph)."New value: +"Does the system perform profiling of natural persons? If true, Art. 6(3) exception is UNAVAILABLE regardless of other conditions (Art. 6(3), third subparagraph)."

Changed1 schema field changedv1.3.0Jun 21, 2026

addedInput schema / properties / no_significant_risk_to_health_safety_fundamental_rights

Added value: +{
+  "description": "Has the provider affirmatively assessed that the system does not pose a significant risk of harm to health, safety, or fundamental rights, including by not materially influencing decision-making outcomes? Required for Art. 6(3).",
+  "type": "boolean"
+}

First observedv1.1.5Jun 14, 2026

TDQS

B3.4/5.0

Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

The description is transparent about being an assessment/walk-through tool, and does not contradict the read-only annotation. It clearly states that profiling disables the exception, which is a critical behavioral aspect. No conflicts with annotations are present.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness2/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is verbose and largely redundant with the schema, repeating all conditions and legal references. While it is structured, it could be condensed significantly without losing essential information, making it less concise than ideal.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness3/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

The description provides important legal context (Art. 6(3), 6(4), 49(2)) and states the documentation and registration requirements. However, it does not explain the tool's output structure or return values, leaving some contextual gaps for a complex assessment tool.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Since the schema provides 100% coverage with detailed per-parameter descriptions, the baseline is 3. The description adds minimal extra nuance, such as the condition for setting 'no_significant_risk', but largely repeats the schema, so it does not elevate the score.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the tool's function: walking through the Art. 6(3) exception for Annex III high-risk AI systems. It specifies the legal basis and the key conditions, making the purpose unambiguous.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description implies usage for assessing whether an Annex III system qualifies for the exception, and mentions documentation requirements. However, it lacks explicit guidance on when to choose this tool over siblings like euaiact_assess_system, and does not provide clear when-not-to-use scenarios.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Deploy Server