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

Annotations already declare readOnlyHint, openWorldHint, idempotentHint, and non-destructive behavior. The description adds useful behavioral context by promising the 'état réel d'application' and listing the specific coverage-failure states it can reveal. No contradiction with annotations.

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

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

Two tight sentences: the first announces the action and scope, the second gives concrete use cases. No filler or redundant restatement of schema information.

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

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

The presence of an output schema covers return structure, the single parameter is fully documented in the schema, and annotations cover behavior. The description supplies the missing contextual layer: why the agent would need this tool and what insights it provides.

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

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

Schema description coverage is 100% and the single optional limit parameter already has an explicit description with default and maximum. The tool description adds no new param detail, so the default baseline of 3 applies.

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

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

The description states a specific verb ('Liste') and resource ('politiques d'accès conditionnel du tenant') and adds a distinguishing property ('état réel d'application'). This clearly separates it from the more user-centric sibling tools such as get_user_signins or get_risk_detections.

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

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

The description explicitly says when to use it: to explain a blocked connection with error code 53003 or to detect coverage gaps caused by disabled policies, audit-only mode, or exclusions. It does not explicitly name alternatives or say when not to use it, but the use cases are strong and concrete.

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

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

Annotations already declare the tool safe (readOnlyHint, idempotentHint, destructiveHint=false). The description adds meaningful behavioral context—that the tool automatically flags persistence or privilege-escalation operations such as role assignments, app secret additions, MFA enrollment, and conditional access changes. No contradiction with annotations.

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

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

The description is compact and front-loaded: the first sentence states the core action and purpose, the second adds high-value behavioral specifics, and the third gives usage guidance. Every sentence earns its place with no redundancy or filler.

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

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

For a tool with only three optional parameters, a rich annotation set, and an output schema, the description is complete. It covers what the tool does, what it highlights, and when to use it. Nothing critical is missing for an agent to select and invoke it correctly.

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

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

Schema description coverage is 100%, so the input schema already documents hours, limit, and initiated_by with defaults and bounds. The description does not add further parameter-level meaning, which matches the baseline of 3 for high schema coverage.

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

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

The description opens with a specific verb and resource: 'Récupère les modifications administratives récentes de l'annuaire'. It also states the investigative purpose (identify configuration drift or unauthorized change), which clearly distinguishes it from sibling tools that focus on user context, sign-ins, or risk detections.

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

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

The final sentence gives an explicit trigger condition: 'À utiliser après avoir constaté une connexion suspecte, pour déterminer ce que l'attaquant a fait une fois entré.' This provides clear context for when to invoke the tool. However, it does not explicitly name alternatives or state when not to use it, so it stops short of a 5.

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

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

Annotations already declare readOnlyHint, idempotentHint, and non-destructive behavior, so the bar is lower. The description adds valuable context beyond annotations by listing risk detection categories and the Entra ID P2 license requirement. It does not over-explain but gives meaningful operational context.

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

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

Three sentences, each earning its place: what the tool returns, how it differs from the sibling, and a hard prerequisite. The content is front-loaded and there is no filler or redundant restatement of the schema.

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

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

Given the rich annotations, full parameter documentation, and output schema, the description covers everything an agent needs: purpose, scope, alternative routing, and licensing. There are no critical gaps for invoking this tool correctly.

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

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

Schema description coverage is 100%, with each parameter already having a clear description and defaults. The tool description does not add parameter-level meaning, but it does not need to; the schema carries the full burden. Baseline 3 is appropriate.

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

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

The description names a specific verb ('Récupère') and a precise resource ('détections de risque unitaires'), and lists concrete examples (identifiants divulgués, IP anonymisée, voyage impossible, pulvérisation de mots de passe). It also explicitly differentiates itself from get_risky_users by explaining WHY versus THAT, making the tool's purpose unmistakable.

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

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

It directly positions the tool against get_risky_users: use this one to understand the reason behind a risk flag, while get_risky_users only indicates the flag exists. It also states a key prerequisite, 'Nécessite une licence Entra ID P2', giving the agent an explicit condition for use.

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

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

Annotations already cover read-only, idempotent, and non-destructive behavior. The description adds useful behavioral context: a required Entra ID P2 license, the risk-ordering behavior, and the scope of accounts returned. No contradiction with annotations.

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

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

The description is compact and front-loaded: main action in the first sentence, use cases in the second, and a prerequisite in the third. Every sentence earns its place with no redundancy.

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

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

Given the simple two-parameter read-only tool, full schema coverage, and an output schema, the description is complete enough. It covers purpose, ordering, use cases, and licensing without needing to duplicate schema information.

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

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

Schema description coverage is 100%; both parameters are already well documented in the schema, including defaults and allowed range. The description does not add additional parameter-level meaning beyond the schema, so baseline 3 applies.

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

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

The description clearly states the tool lists accounts flagged as risky by Entra Identity Protection, sorted from most to least risky. It is specific about the resource and action, though it does not explicitly name sibling tools for differentiation.

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

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

The description explicitly says when to use it: to get an overview of tenant exposure or to confirm a specific account is considered compromised. It provides clear usage context, but does not mention exclusions or alternative sibling tools.

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

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

Annotations already declare readOnlyHint, idempotentHint, and destructiveHint false, so the safety profile is covered. The description adds the context that this tool is authoritative for severity and that it returns identity attributes, but it doesn't disclose additional behavioral details such as staleness, data-source scope, or potential edge cases. With annotations handling the core safety traits, this is adequate but not rich.

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

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

The description is front-loaded with the action and resource, then adds two purposeful sentences about severity and systematic use. It is concise and every sentence contributes meaningful guidance. The only minor issue is that the severity phrasing is slightly redundant with the usage instruction, but this is not wasteful.

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

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

For a single-parameter tool with a rich output schema and strong annotations, the description is complete: it explains what the tool returns, why it matters, and exactly when to invoke it. Nothing essential is missing for an agent to select and call it correctly.

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

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

Schema description coverage is 100%, and the upn parameter already has a clear explanation with an example. The description adds no extra parameter semantics, but the baseline of 3 applies because the schema does the heavy lifting.

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

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

The description states a specific verb and resource: 'Récupère la fiche d'identité d'un compte' and enumerates the exact fields returned (poste, département, état, groupes, rôles). It also distinguishes itself from siblings by framing the tool as the one that determines incident severity, so an agent can clearly tell it apart from sign-in or risk-detection tools.

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

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

The description gives strong usage direction: it should be called 'systématiquement avant de conclure sur un incident' and explains why severity depends on account privileges. It lacks explicit when-not-to-use guidance or named alternativatives, so it doesn't quite reach a 5, but the context is clear.

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

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

Beyond the readOnly/idempotent annotations, the description discloses that events are reduced to security indicators and that a quantitative summary (failure counts, distinct IPs, computed observations) is returned. This is useful behavioral context, though it does not discuss rate limits or response edge cases.

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

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

The purpose and output behavior are front-loaded and each substantive sentence earns its place. The Args block is somewhat redundant with the input schema, but it is compact and does not make the description overly long.

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

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

With a rich output schema and annotations, the description provides enough context for correct selection and invocation: purpose, use cases, output shape, and parameter defaults. It could be slightly more complete by pointing to a sibling tool for related risk data, but this is not essential.

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

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

The schema already documents all three parameters, including defaults and bounds, so coverage is 100%. The description largely repeats this information (upn example, hours default/max, limit default/max) and adds no new parameter semantics beyond the schema.

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

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

The description clearly states the tool retrieves recent user sign-ins ('Récupère les connexions récentes d'un utilisateur') for authentication-block, geographic-anomaly, or compromise investigations. It uses a specific verb+resource, but does not explicitly contrast itself with siblings such as get_risk_detections or get_user_context.

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

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

It gives concrete investigative scenarios: 'pour investiguer un blocage d'authentification, une anomalie géographique ou une compromission'. There is no explicit when-not-to-use guidance or named alternative, so it stops short of a 5.

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