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

Far exceeds the annotations: it discloses that queued/preparing tasks will not dispatch images, that running tasks only stop local waiting and are re-marked 'unknown', and that remote stop/refund is not guaranteed. This is exactly the caveat an agent needs before invoking a cancel, and it is consistent with idempotentHint=true and destructiveHint=false.

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?

A single dense sentence, front-loaded with the action and then the per-state consequences; every clause carries information. The semicolon-separated structure is slightly compressed but readable and waste-free.

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?

Behavioral consequences are well covered and an output schema exists so return values need no explanation, but the definition leaves the three required parameters (especially checkpointer/thread_id) unexplained. For a 3-required-param mutation tool, that is a meaningful completeness 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 description coverage is 0% for all three required parameters. The description only vaguely implies task_ids ('指定任务') and says nothing about what checkpointer or thread_id mean or how to obtain them, so it fails to compensate for the schema 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 names a specific verb+resource ('取消指定任务' / cancel specified tasks) and immediately qualifies behavior by task state (queued/preparing vs running). It is clear what the tool does, though it never names a sibling tool to differentiate itself from alternatives like get_image_tasks or submit_image_tasks.

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 implicitly guides usage by explaining the effect in each task state, which helps an agent decide whether cancelling is worthwhile. However, it gives no explicit when-to-use/when-not-to-use framing and never points to an alternative tool for related needs (e.g., checking status before cancelling).

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 readOnly/idempotent/openWorld=false, yet the description adds genuine behavior: it does not call upstream, does not resend requests, and returns isError for incomplete/failed batches. The isError error-handling semantics are real value beyond the structured hints.

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 clauses with the core action front-loaded and the behavioral caveats following; no wasted wording, though it is terse to the point of omitting parameter context.

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 1-param read tool with annotations and an output schema, the safety and error-handling picture is adequately covered. The only real gap is the unexplained request_id, but return values need not be described given the output schema.

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% and the single request_id parameter is never explained in the description — no format, source, or meaning ('已登记批次' only loosely implies an id). The description does not compensate for the schema 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?

Names a specific verb+resource: reads an already-registered batch and its save path (读取已登记批次及保存路径). This is discernible from the sibling task-submission tools, though it does not explicitly contrast itself with get_image_tasks.

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?

Implies usage via the constraints '不调用上游、不重发请求' — i.e. use this to fetch a result without re-triggering work — but it never states when to prefer it over get_image_tasks or under what conditions a batch is 'registered'.

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?

Goes beyond the readOnly/idempotent annotations by disclosing status-dependent return behavior: queued/preparing/running succeed, while failed/unknown/cancelled surface as isError with full status preserved. It also warns that checkpointer is a SQLite namespace, not a path or credential. Missing pagination/return-shape details keeps it at 4.

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 packed sentences, front-loaded with the core action and contrasted against resubmission, followed by status behavior and the checkpointer gotcha. Little waste, though the checkpointer sentence could be trimmed.

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?

An output schema exists, so return values need not be described; the description instead covers error semantics and the checkpointer pitfall, which are the real risks for a 5-param retrieval tool. The undocumented thread_id/limit and absent alternative-tool routing are the remaining 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%, so the description must carry parameter meaning. It clarifies task_ids (filtering), pagination, and especially checkpointer semantics, which is high-value. However thread_id and limit go unexplained, and it says 'next_cursor' while the schema param is 'cursor', so the coverage is only partial.

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 (查询/query) and resource (持久任务与图片绝对路径 – persistent tasks plus image absolute paths), and explicitly contrasts with the write sibling via '不重新提交上游'. It doesn't fully disentangle itself from get_batch_result, so it stops 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?

Gives concrete usage context: filter by task_ids or paginate with the cursor, and it signals a when-not relative to submit_image_tasks ('does not resubmit upstream'). No explicit guidance on choosing between this and get_batch_result, so not 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=false, and the description adds useful behavioral detail: it explicitly says the tool does not connect to the network and does not return keys. This is meaningful transparency beyond the structured annotations, though it does not elaborate on caching, freshness, or return shape.

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 a single compact sentence that front-loads the main purpose and appends key constraints after a semicolon. Every clause carries useful information with no redundant restatement of the tool name.

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 zero parameters and an output schema present, the description does not need to explain return values in detail. It covers the resource being listed and the important behavioral constraints, making the definition complete enough for an agent to 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?

The tool has zero parameters, so the baseline for parameter semantics is 4. There is no parameter behavior to clarify, and the description appropriately does not invent any parameter semantics.

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 uses a specific verb (列出/list) and enumerates the exact resource contents: seven models with 2k/4k variants, API formats, access directories, and restrictions. This clearly distinguishes it from the sibling image-task tools, which submit, query, cancel, or batch-process images.

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 explains what is returned but gives no explicit guidance on when to use this tool, when not to use it, or which sibling to choose instead. The use case is implied by the name and contents, but no actual usage context or prerequisite is stated.

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=false, so the safety profile is covered. The description adds genuine context beyond that: the operation is free, produces a reviewable plan, never fabricates product info, and defers reference-image download/validation until actual generation. That is meaningful behavioral disclosure for a planning tool.

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?

Purpose is front-loaded in the first line, followed by a tight per-parameter glossary with no padding. Each line maps to a parameter, though the glossary style is terse rather than fully integrated prose.

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?

An output schema exists so return values need no explanation, and the description covers the planning-specific behavior and key params. Gaps remain: no workflow relationship to submit_image_tasks/list_models and four parameters are unexplained, leaving the definition adequate but incomplete for a complex 9-param 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?

With 0% schema coverage the description must carry parameter meaning, and it does explain scene enum values, selling_points semantics, count guidance, reference_images handling and resolution routing. But it leaves aspect_ratio, copy_text, requirements and product_name completely undocumented, so roughly half the surface is still opaque.

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 (生成 plan) and resource (主图/详情页/场景/海报绘图计划), and the phrase 可审阅的...计划 clearly distinguishes it from the sibling submit_image_tasks that actually produces images. It stops short of naming any sibling explicitly, but the plan-vs-execute distinction is legible.

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 note that reference images are only read/downloaded '真正出图前' implies this is a pre-generation planning step, and '不自动编造商品信息' hints at the intended input discipline. However it never states when to call this versus submit_image_tasks, nor any ordering or prerequisite.

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 annotations it discloses that billing happens in the background, that task_id is unique/stable within checkpointer/thread_id with content reuse, and critically that any conflict aborts the entire batch. It also explains restart semantics (unsent tasks resume, possibly-sent tasks do not auto-retry). This is exactly the extra behavioral context annotations cannot carry.

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 most important behavior (immediate return, no waiting) is front-loaded, and each subsequent line carries distinct information. It is dense and telegraphic but nearly every clause earns its place; a little more structure would have made it cleaner.

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 complex nested batch-submission tool with an output schema (so return values need not be described), the definition covers idempotency, conflict behavior, billing, model routing, status interpretation, and restart semantics. Nothing an agent needs to invoke it safely 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?

Top-level schema coverage is 0%, so the description must compensate, and it largely does: it explains task_id uniqueness/conflict semantics and model/api_format/endpoint defaults, and adds routing rules (Flash-lite only Gemini/Chat, nano-banana-pro only DALL·E, gemini-nano only DALL·E Generations/Gemini/Chat) plus resolution-based variant selection. It does not fully decode every nested field, so not a 5.

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+scope: batch-registers 1–12 background image tasks and returns immediately without waiting for rendering/download. It distinguishes itself from get_image_tasks by explicitly naming it as the query counterpart, so an agent can route correctly 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?

It gives clear conditional guidance: use get_image_tasks to query, treat queued/preparing/running as normal, and do not blindly resubmit on 'unknown'. It also supplies model-routing constraints. It doesn't explicitly compare against plan_ecommerce_images or cancel_image_tasks, so it stops short of a full when/when-not map.

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