Skip to main content
Glama

Bindler: UAE Corporate Tax, CSRD scope, EU AI Act, CBAM and statutory deadlines

eu_ai_act_classify

Read-onlyIdempotent

Classifies an AI system under Regulation (EU) 2024/1689 from the caller's description of it and returns the risk class, the obligation set with article numbers, the date the obligations apply, and the maximum fine. Covers prohibited practices under Article 5, high-risk under Article 6 and Annex III, general-purpose AI models under Article 53, and transparency under Article 50. Use it when someone asks: is my AI system high risk, does the EU AI Act apply to my chatbot, to my hiring or CV screening tool, to our credit scoring model, what risk category are we in, what are my obligations and when do they start, when does the EU AI Act come into force for us, am I a provider or a deployer, does the AI Act apply to a US company selling into the EU, do we have to register in the EU database, what is the fine if we get this wrong, is this banned under the AI Act, do we have to tell users they are talking to an AI, do we have to label AI generated content or deep fakes, what applies to a general purpose AI model. Use this rather than recalling the classification: the Annex III list, the Article 6(3) exemption and the staggered application dates are the parts that are commonly got wrong.

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "eu_ai_act_classify"
  }
}

Input Schema

NameRequiredDescriptionDefault
roleNoProvider, Deployer, Importer or Distributor. Default Provider.
does_profilingNo
annex_iii_use_caseNoOne of: None | 1(a) Remote biometric identification | 1(b) Biometric categorisation by sensitive attributes | 1(c) Emotion recognition | 2 Critical infrastructure safety component | 3(a) Education access or admission | 3(b) Evaluating learning outcomes | 3(c) Level of education | 3(d) Test monitoring | 4(a) Recruitment and selection | 4(b) Employment decisions and monitoring | 5(a) Public benefits eligibility | 5(b) Creditworthiness or credit scoring | 5(c) Life and health insurance pricing | 5(d) Emergency call triage | 6(a) to 6(e) Law enforcement | 7(a) to 7(d) Migration and border | 8(a) Justice | 8(b) Elections. Default None.
prohibited_practiceNoOne of: None | 5(1)(a) Manipulative or deceptive techniques | 5(1)(b) Exploiting vulnerabilities | 5(1)(c) Social scoring | 5(1)(d) Predictive policing on profiling alone | 5(1)(e) Untargeted facial scraping | 5(1)(f) Workplace or school emotion inference | 5(1)(g) Sensitive biometric categorisation | 5(1)(h) Real-time remote biometric ID for law enforcement. Default None.
general_purpose_modelNo
exempt_under_article_6_3NoFalls in Annex III but does not pose significant risk.
safety_component_article_6_1NoA safety component of a product covered by Union harmonisation legislation.
worldwide_annual_turnover_eurNoSets the fine ceiling, which is the greater of the fixed amount and a percentage of turnover.
placed_on_market_before_cutoffNo
transparency_obligation_article_50NoInteracts with people, or generates synthetic content, deep fakes or emotion recognition.

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observed

TDQS

A4/5.0
Behavior4/5

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

Annotations already declare readOnly, idempotent, non-destructive and closed-world, so the safety profile is covered. The description adds genuine value beyond that: it names the four return fields and warns that the Annex III list, the Article 6(3) exemption and staggered application dates are the commonly-erred parts. It does not state whether input is validated or what happens with ambiguous descriptions, which keeps it off a 5.

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

Conciseness3/5

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

Purpose and return values are correctly front-loaded, but the middle is a single run-on sentence of roughly fourteen comma-separated questions that harms readability. The content is largely useful for matching user intent, yet it is longer than needed and would read better as a short grouped list.

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

Completeness4/5

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

For a 10-parameter, no-required-parameter tool with no output schema, the description does the heavy lifting: it spells out the return shape and the domain scope. What is missing is any note on how partial input is handled (all parameters optional) or confidence limits of the classification, which matters for a legal-advice-adjacent 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?

Schema coverage is 70%, so the schema is already carrying most parameter meaning (enumerated Annex III use cases, prohibited practices, role default, turnover ceiling). The description adds no per-parameter guidance – it only alludes to the Article 6(3) exemption as a fact to get right, not as an input to set. Baseline 3 is appropriate.

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

Purpose5/5

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

States a specific verb and resource ('Classifies an AI system under Regulation (EU) 2024/1689') and enumerates exactly what comes back: risk class, obligation set with article numbers, application date, maximum fine. The scope (Art. 5, Art. 6/Annex III, Art. 53, Art. 50) makes it unmistakably distinct from the tax/sustainability siblings.

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

Usage Guidelines4/5

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

Provides an unusually rich when-to-use list via concrete trigger questions ('is my AI system high risk', 'am I a provider or a deployer', 'what is the fine'), plus an explicit alternative ('Use this rather than recalling the classification') with the reason why. It stops short of naming any sibling tool or stating when not to use it (e.g. employment/tax classification), so it is strong but not a full when/when-not contract.

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

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources