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

Annotations already declare readOnlyHint=true and openWorldHint=true, so the safety profile is covered. The description adds valuable behavioral context the annotations can't: the host model retains final authority, and output should not be mechanically followed. It omits latency/cost/determinism of the external call.

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 dense sentences with zero waste: the usage condition and the authority caveat are front-loaded, and the exclusion ('open-ended coding/prose') closes the second sentence. Every clause earns its place.

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 read-only advisory tool with no output schema, the description covers invocation conditions, boundaries, and authority semantics well. It does not describe the shape of the response or what the model override does, which leaves a small gap given no output schema exists to carry that.

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 coverage is 75%, so the schema already documents model, state, and question. The description's '2-255 explicit alternatives' maps meaning onto the options array and its bounds, but adds nothing for state or question beyond the schema. 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 states a specific capability: producing an independent second opinion on a bounded decision with 2-255 explicit alternatives. It does not name its sibling tools (jev_gate, jev_route, jev_risk_score), so differentiation relies on the 'bounded decision' framing rather than explicit routing.

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?

Strong when-to-use ('only after inspecting enough context when a bounded decision has 2-255 explicit alternatives and genuine uncertainty remains') and when-not-to-use ('do not... use it for open-ended coding/prose'). It doesn't explicitly reference the sibling tools as alternatives, which keeps it from 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=true and openWorldHint=true, so the safety profile is partly covered. The description adds substantive context beyond that: the tool is explicitly 'advisory' and its output is not authorization for irreversible or security-sensitive actions, which is important behavioral framing an agent needs before acting on a pass. It still does not say what a pass or fail returns, what the default 0.8 threshold implies, or what happens on a failed gate.

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 sentences, front-loaded with the operational trigger followed immediately by the critical caveat. Every clause carries weight and nothing is redundant.

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 4-parameter tool with 0% schema description coverage and no output schema, the description leaves the central contract unstated: what 'state' and 'statement' are, what a gate pass/fail actually returns, and how the threshold maps to the outcome. The advisory/not-authorization framing is valuable but does not compensate for those gaps.

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 0% across 4 parameters, so the description must carry the parameter burden and largely does not. It gestures at 'probability threshold' (mapping loosely to the threshold parameter) but never explains what state, statement, or model should contain, nor the 0-1 range and 0.8 default.

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 capability: an atomic yes/no gate that acts as an advisory second opinion, conditioned on a probability threshold. An agent can grasp what the tool does. It does not, however, differentiate itself from the sibling tools (jev_decide, jev_route, jev_risk_score), which all sound like judgment/decision primitives.

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 a real selection condition ('when a probability threshold would materially change the next action') and a genuine exclusion (passed gates are not authorization for destructive, irreversible, production, security, or data-loss-sensitive actions). It never names an alternative sibling or says when one of those should be preferred instead, so it stops short of full routing guidance.

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 mark it read-only and open-world; the description adds non-obvious behavioral limits: it is advisory only and explicitly subordinate to tests, evidence, reasoning, and review. It also states return categories, but does not describe model/state handling or output format beyond the category list.

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 tightly constructed sentences, front-loaded with the usage directive and return categories, then the critical limitation. No 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?

Purpose, usage boundaries, and return categories are covered, which matters given the absence of an output schema. However, with 33% schema coverage and optional model/question parameters, the description leaves parameter-level guidance to the schema and does not compensate for the gaps.

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 only 33%: state has a description, model and question do not. The description does not compensate by explaining these parameters, the anyOf state formats, or question semantics, so it adds little 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?

States the tool returns a risk score with four categories and frames it as an advisory second opinion. Its role is distinguishable from decision/gate siblings by the non-authorization and non-override framing, though no sibling is named explicitly.

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?

Specifies when to use it: for consequential or ambiguous-risk changes after inspecting context. It also sets when-not boundaries: never authorization and never overrides tests, deterministic evidence, primary-model reasoning, or review. Alternatives are implied but not named.

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=true and openWorldHint=true, and the description is consistent with them. It adds one meaningful behavioral constraint — "Never changes the user-selected host model" — which clarifies the tool is advisory only. It says nothing about the shape of the result or latency, so the added context is modest.

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 usage condition and the candidate scope are front-loaded. No filler or redundancy, though the second sentence about the host model is somewhat tangential to selection.

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 4-parameter, 3-required tool with no output schema and half its parameters undocumented, the description covers the decision context but not the return value or what "second opinion" output looks like. Adequate but with clear gaps.

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 only 50%: state, objective, and candidates are documented in the schema, but the model parameter has no description in either place. The description text adds no parameter-level meaning at all, so it fails to compensate for the coverage gap.

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 function — producing a routing "second opinion" over a fixed candidate set (tools, subsystems, workstreams, handling paths) — which is concrete enough to distinguish it from a scoring or gating tool. It does not explicitly name or contrast with its siblings (jev_gate, jev_decide, jev_risk_score), so it falls short of a 5.

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 an explicit when-to-use condition: only for uncertain routing where deterministic evidence does not already decide the route. That is a real exclusion criterion. However, it never names the alternative tool to use when the route IS deterministic, so it stops at a 4.

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