VerifyAX MCP

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

Annotations already declare destructiveHint=true and idempotentHint=true, so safety is covered structurally. The description adds genuine value beyond that by warning the deletion is permanent and cannot be undone, and by noting a confirmation is returned.

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 short, front-loaded sentences with no filler; each adds a distinct fact (what, irreversibility, return confirmation).

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 destructive tool with annotations covering safety and no output schema, the description supplies the irreversibility warning and return confirmation, leaving little an agent needs missing. Minor gap: no permissions or preconditions context.

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?

Only one parameter, and schema description coverage is 100% with the uuid semantics fully documented in the schema. The description merely restates that the agent is identified by uuid, adding no syntax or format detail, so the 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?

States a specific verb (deletes) and resource (agent) with the identifying key (uuid) and workspace scope. It is distinguishable from delete_scenario by the named resource, though it never explicitly contrasts with that sibling.

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?

No guidance on when to use this versus alternatives such as list_agents or register_agent, and no prerequisites stated (e.g. permissions, required prior state). The agent must infer the calling context entirely.

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 destructiveHint=true and idempotentHint=true, yet the description still adds meaningful context: the deletion is permanent and cannot be undone, and it is blocked while simulation runs reference the scenario. The only omission is confirmation of return shape, which is mentioned but not detailed.

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 tight sentences, front-loaded with the action and resource, then the irreversibility/constraint, then the response. 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?

For a single-parameter destructive tool, the description covers permanence, the referential failure mode, and the return outcome. Nothing an agent needs before invoking it is missing.

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?

One parameter with 100% schema description coverage; the schema already documents scenario_uuid fully. The description only restates that the uuid identifies the scenario, adding no format or sourcing detail 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 a specific verb (permanently deletes), resource (scenario), and scope (by uuid, in the VerifyAX workspace). It is clearly distinguishable from sibling delete_agent and from read/list 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?

Gives a clear precondition for use and failure: it fails if simulation runs still reference the scenario. It does not name an alternative tool or spell out when-not-to-use beyond that precondition, but the context is unambiguous.

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 only declare readOnlyHint=false and destructiveHint=false, so the description carries the real behavioral load and delivers: 30s–30min runtime, pollable task handle for task-capable clients vs blocking behavior for others, cost preview before the run, and the wait-then-return-scores flow. This is exactly the operational context an agent needs before committing to a long 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?

Three sentences, front-loaded with the action and duration, followed by client-behavior detail and input requirements. It is tight and well-ordered; the 'give it an agent uuid and a scenario uuid' clause mildly duplicates the schema but reads as a necessary lead-in to the workflow.

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 long-running tool with no output schema, the description still conveys the execution model (async task handle vs blocking), expected latency, cost preview, and the fact that evaluation scores are returned. Nothing an agent needs to invoke it correctly is missing.

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 all four parameters are already documented, including num_runs and the 1–240 range on timeout_minutes. The description's mention of timeout_minutes overriding the scenario default restates what the schema already says, adding no new syntax or format detail. Baseline 3 applies when 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+resource chain ('runs an agent against a scenario and evaluates the result end to end') and enumerates what it does (preview cost, run simulation, wait, return scores). It is clearly distinct from get_run_details or preview_run_cost by implication, but it never names a sibling or explicitly excludes one, so it stops short of full sibling 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?

Usage is implied rather than stated: the agent is told what inputs to supply ('give it an agent uuid and a scenario uuid') but not when to choose this tool over preview_run_cost or get_run_details. No exclusions or prerequisites are given, leaving routing to inference.

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 declare only that this is not read-only and not destructive. The description adds substantial context beyond that: validation failures return 400/422 synchronously before any job is queued, typical latency is 30s–2min, task-capable clients get a pollable handle while others block, and post-start failures return a structured error.

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?

Front-loads the core purpose and validation behavior, then layers timing, async semantics, batch mode, and return values in a logical order. Each sentence carries information, though the density of caveats makes it heavier than ideal.

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?

There is no output schema, and the description compensates by describing the return value (scenario uuid, batch uuids, or structured error). Combined with validation behavior, latency, and async handling, an agent has everything needed to call this 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 coverage is 78%, so most parameters are documented structurally, but the description adds cross-parameter semantics the schema does not: num_scenarios > 1 requires tag_pool, and that mode is called "batch mode". It does not explain total_tags, include_tags, or max_tags_per_npc beyond what the schema already states, so it stops short of full compensation.

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?

Opens with a specific verb+resource ("Generates a new test scenario") and immediately enumerates the scenario_type values (info_exchange or interview). It clearly differentiates itself from the sibling tag-discovery tools by naming them as inputs to this operation rather than alternatives to it.

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?

Explicitly routes the agent: use recommend_scenario_tags, search_scenario_tags, or list_compatible_tags to pick valid tag names. It states the condition for batch mode (num_scenarios > 1 requires tag_pool) and explains the client-dependent blocking vs task-handle behavior, so an agent knows exactly what to expect.

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?

readOnlyHint=true already tells the agent this is a safe read, so the bar is lower. The description adds real value by disclosing that evaluation results are conditional ('when they are available'), warning the agent that scores may be absent for an in-progress run. It says nothing about error behavior for bad/unknown uuids.

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 with the purpose front-loaded and the usage note second. The second sentence mildly restates the first ('inspect scores and outcome' vs 'status and the evaluation results'), a small redundancy but not bloat.

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?

There is no output schema, so the description must carry the return-value burden, and it does name the key fields (status, evaluation results) plus their conditional availability. For a one-param read tool this is essentially sufficient; only uuid error handling is unaddressed.

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 100% and there is a single parameter whose description ('The run's uuid.') matches what the tool description conveys. The description adds no format, constraint, or lookup guidance beyond the schema, so the 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?

States a specific verb ('Fetches') and resource ('full details of a single simulation run'), with the identifier ('by its uuid') and expected payload (status, evaluation results). It implicitly separates itself from the list_recent_runs sibling by scoping to a single run, but never names an alternative 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?

'Use after a run to inspect scores and outcome' gives clear timing and intent for when to reach for this tool. It stops short of naming the alternative (list_recent_runs) or stating when not to use it, so it is context without 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?

readOnlyHint=true already tells the agent this is a safe read. The description adds a genuinely useful behavioral detail that annotations do not cover: pagination across all matching events up to a cap, and that spend is returned only 'when the API reports it' – an important conditional the agent should know about. It still omits how the cap interacts with max_events and whether partial results are signaled.

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 clauses in a single tight sentence, front-loading the summary action, then scoping, then return contents. No filler or repetition.

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 10-parameter read-only aggregator with no output schema, the description covers purpose, scope, pagination, and the shape of the return value well. It stops short of explaining the four undescribed parameters or the difference between the various *_uuid filters, which is the main remaining gap.

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 60%, so half the parameters (failed, product_area, scenario_uuid, simulation_uuid) have no inline descriptions. The prose gestures at the filter dimensions but does not explain semantics for the undocumented parameters, leaving the description to partially compensate rather than fully cover the gap. 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?

States a clear verb+resource (summarizes VerifyAX usage events) and even names the output fields (event count, breakdown by product area, platform spend). It is not explicitly differentiated from the nearest sibling preview_run_cost, but the purpose is unambiguous.

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 mentions the filter dimensions obliquely ('over an optional time range or for a specific simulation, scenario, or job') but never states when to use this tool versus preview_run_cost or other cost-related siblings, nor any prerequisites or exclusions.

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?

readOnlyHint=true already tells the agent this is a safe read, so the description's burden is lower. It adds useful context by enumerating the filter values and the returned fields (uuid, name, type, URL), but says nothing about pagination behavior or result ordering despite exposing limit/offset.

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 core action and optional filter come first, the return shape second. No filler, no repetition beyond the necessary enum listing.

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?

Since there is no output schema, the description correctly covers the return fields, which is exactly what an agent needs. The remaining gap is pagination semantics for limit/offset, the only substantive omission for a list tool.

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 only 33% – agent_type is documented in the schema, while limit and offset have no description anywhere. The description merely restates the enum values for agent_type, an optional filter, and never explains the pagination parameters, so it fails to compensate for the documentation 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 gives a specific verb and resource ("Lists the AI agents registered in your VerifyAX workspace") plus the filter dimension, so it is unambiguous what the tool returns. It does not explicitly contrast itself with siblings like register_agent or delete_agent, but the resource is distinct enough that confusion is unlikely.

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 notes the filter is optional and names the allowed connector types, which implies the use case (browsing/filtering registered agents). However, it never states when to prefer this over other discovery tools or what happens when no filter is supplied, leaving usage to inference.

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?

readOnlyHint=true already tells the agent this is a safe read. The description goes further by disclosing the return contents (name, category, description) and a domain-specific constraint — QnA tags flagged as needing to be the only tag — which is behavioral context not present in the annotations or schema.

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: purpose and filter up front, then the alternative-tool routing, then the return summary. No filler, and the most decision-relevant information (what it returns, which sibling to use) leads.

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?

No output schema exists, yet the description covers the return shape and the special QnA flag, and an annotation covers the safety profile. For a one-parameter read-only listing tool, nothing needed to call it correctly is missing.

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 enum parameter is fully documented, so the baseline is 3. The description restates the two enum values (info_exchange, interview) but adds no syntax, format, or defaulting detail beyond what the schema already provides.

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 a specific verb (Lists) and resource (full skill-tag catalogue) with explicit scope: filtered to tags compatible with a given scenario type. It also names the sibling tools it is not (recommend_scenario_tags, search_scenario_tags), so an agent can distinguish it without opening schemas.

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?

Gives an explicit alternative and the selecting condition: use recommend_scenario_tags or search_scenario_tags for a ranked shortlist when the user gave natural language or the list is too large. This is a clear when-to-use-this-vs-others rule.

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?

readOnlyHint=true already declares the safety profile, so the description's main added value is naming the returned fields (uuid, status, agent, scenario, evaluation handle). It does not mention pagination behavior despite limit/offset, and adds no rate-limit or ordering 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?

Two tightly written sentences with no filler; the purpose is front-loaded and the filter list and return summary follow logically.

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 no output schema, the description does the right thing by enumerating the returned fields. The only material gap is the absence of any pagination (limit/offset) or result-ordering note for what is clearly a paged listing tool.

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 56%, so roughly half the parameters carry no inline documentation. The description compensates by naming the filterable dimensions (status, agent, scenario, date range, search text, run group), matching the schema fields but adding no syntax or format detail beyond them.

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 a specific verb (Lists) and resource (recent simulation runs) scoped to a workspace, and enumerates the available filter dimensions. An agent can distinguish this bulk-list tool from the singular get_run_details sibling without opening either schema.

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 word 'optionally filtered' implies this is a browse/query tool, but it never states when to reach for it versus get_run_details or how the filters combine. Usage is implied rather than directed.

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?

The readOnlyHint annotation already covers the safety profile, so the description earns credit for disclosing the return fields (uuid, name, type, status). However it says nothing about pagination behavior despite limit/offset existing, leaving result-set handling opaque.

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 with zero waste. The scope and filters are front-loaded and the return shape follows immediately, so every sentence 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 list tool with no output schema, the description usefully names the returned fields. It is nearly complete, with the only gap being the unexplained limit/offset paging semantics.

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 25%: status is documented, scenario_type has an enum, but limit and offset are bare in both schema and description. The description does reinforce the type values (info_exchange/interview) and the status filter, but does not compensate for the undocumented pagination parameters.

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 a specific verb ('Lists') and resource ('test scenarios') scoped to the VerifyAX workspace, and enumerates the supported filters. It is clear and self-contained, though it doesn't explicitly contrast itself with sibling list/search tools like search_scenario_tags.

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?

'optionally filtered by type ... and status' implies the usage context, but there is no explicit when-to-use vs when-not guidance and no alternative tool named. The agent must infer that this is the unfiltered enumeration tool rather than a search sibling.

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?

The readOnlyHint annotation already covers the safety profile, so the description's added value is the disclosure of what is returned: estimated credits, current balance, and pending committed spend. It also notes that timeout_minutes feeds into the estimate. It stops short of stating whether estimates are guaranteed or how stale the balance figure is.

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 tight sentences: purpose and timing first, then return payload, then the one parameter worth calling out. No filler, no repetition of the name or title, and the most decision-relevant clause is front-loaded.

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?

There is no output schema, and the description compensates by enumerating the three return values. Combined with a complete parameter schema, full annotation coverage of the read-only nature, and the stated pre-run timing, an agent has everything needed to call this 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 baseline is 3 per the rubric. The description restates timeout_minutes and its 1-240 range and adds the causal note that it 'affects the run-cost estimate', which is marginally useful, but it adds nothing about num_runs or agent_uuid beyond what the schema already documents.

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 a specific verb and resource ('Estimates the credit cost of running an agent against a scenario') plus the key temporal qualifier 'before triggering it'. No sibling tool (evaluate_agent, get_run_details, get_usage_summary) performs pre-flight cost estimation, so the agent can distinguish it immediately.

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?

'before triggering it' clearly establishes the pre-flight use context, which is the main decision point for an agent choosing this tool. However, it names no alternative and gives no explicit when-not condition (e.g., that this is read-only and safe to call speculatively), so it falls 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 only declare readOnlyHint, and the description adds real context: 'Read-only (no generation credits)', the embedding+LLM pipeline, a ~20 tag cap, and timeout/unavailability behavior. It stops short of describing latency or the exact ranking semantics, so it is strong but not exhaustive.

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?

Front-loaded with purpose then routing then prerequisites then fallbacks; every sentence carries information. It is dense and slightly over-stuffed for a 3-parameter tool, but 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?

No output schema exists, and the description does convey the return shape (ranked shortlist of up to ~20 tags) plus failure handling, so an agent can call and recover correctly. Minor gaps remain around what happens when neither optional param yields a match.

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 100%, so baseline is 3, but the description adds a constraint the schema does not encode: 'Provide context_prompt and/or agent_uuid (at least one)', which clarifies that scenario_type alone is technically valid but semantically insufficient.

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 a specific verb+resource ('Suggests skill tags for a scenario type') and names the mechanism (Workbench pipeline, embeddings plus LLM, up to ~20 tags). It is immediately distinguishable from siblings like list_compatible_tags and search_scenario_tags.

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?

Explicit when-to-use ('user describes a scenario or agent and wants a ranked shortlist before generate_scenario'), explicit preference over list_compatible_tags for large catalogues or natural-language context, and explicit fallbacks on timeout ('try search_scenario_tags or list_compatible_tags'). This is a complete routing guide.

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 only declare readOnlyHint=false and destructiveHint=false, so the description carries the interesting behavior and does so well: it discloses the pre-creation reachability check for A2A and the probing for DIRECTLINE/MCP, states the return payload (new uuid plus whether connectivity was checked), and warns that credentials transiting the conversation should be supplied out of band. That is meaningful side-effect and security context beyond the 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?

Four sentences, front-loaded with the core action and connector enumeration before the behavioral and security notes. Dense but each sentence contributes; only slightly longer than strictly necessary.

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 17-parameter mutation tool with no output schema, the description supplies the return values, connector-specific behavior, and a credential-handling caveat, which covers the agent's main decision points. Remaining gaps are the undocumented optional tuning parameters, which the 35% schema coverage does not fill in either.

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 35% across 17 parameters, so the description is expected to compensate. It identifies the primary inputs (name, connector type with its enum values, URL, optional auth/connector settings), which covers the essential fields, but leaves many optional parameters (timeout, include_full_context, max_requests_per_minute, default_output_modes, agent_card_path, etc.) unexplained in either place.

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 (registers) and resource (AI agent in VerifyAX), then enumerates the four connector types (A2A, API, DIRECTLINE, MCP) that shape the call. This cleanly distinguishes it from the read/delete siblings such as list_agents and delete_agent without the agent needing to open any schema.

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 explains what happens per connector type (A2A card verification, DIRECTLINE/MCP probing) but never states when to prefer this tool over alternatives or any prerequisites/exclusions. Usage is implied by the connector-type branching rather than made explicit.

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 the description confirms read-only, so no contradiction. It adds genuinely useful behavior beyond annotations: a per-user tag-search rate limit on top of the workspace API limit, plus concrete handling ('honor Retry-After, wait and retry'). It stops short of describing the return shape, but the rate-limit disclosure is above the bar. Note: no output schema exists, so return details would have been a bonus.

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 tight sentences, front-loaded with what it does and how it compares to the sibling, then the operational caveat. Every sentence earns its place, though the rate-limit clause is packed densely.

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 search tool with a fully described schema, the definition covers purpose, alternatives, and the rate-limit failure path, which is what an agent needs to invoke it correctly. The only gap is return-shape detail, minor because results are implicitly tag names and no output schema is declared.

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 all three parameters (scenario_type enum, query natural-language text, limit) are already documented in the schema. The description adds no syntax or format detail beyond what the schema provides, 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?

States a specific verb+resource (finds skill tags), the mechanism (embedding similarity to a natural-language query, no LLM), and the scoping dimension (scenario type). It also distinguishes itself from the sibling recommend_scenario_tags by cost, so an agent can tell the two apart without opening schemas.

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?

Explicitly names when to use it ('find tags about X', cheaper than recommend_scenario_tags) and states the fallback for the failure case ('use list_compatible_tags for the full filtered catalogue'). Alternatives and conditions are spelled out rather than left to inference.

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