AI 模型下架與替代日程

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

With no annotations, the description carries the full behavioral burden but only implies a read-only listing. It does not state what happens when month is omitted (the parameter is optional), whether the result is a calendar of all known dates, or how it relates to the 'upcoming' sibling.

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 short sentence with the scope front-loaded and no filler. It is efficient, though arguably under-specified rather than maximally informative.

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 annotations and no output schema, the description must resolve ambiguity about the optional month parameter and the overlap with upcoming_ai_model_shutdowns; it does neither, leaving an agent to guess the correct tool and call pattern.

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 the YYYY-MM format for month is already documented in the schema. The description adds the notion of grouping by month but does not address the fact that month is optional or what the default range is.

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 (列出/list) and resource (模型下架日/model shutdown dates) with a scoping modifier (依月份/by month), so the operation is identifiable. It does not, however, differentiate itself from the sibling upcoming_ai_model_shutdowns, which appears to cover overlapping subject matter.

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?

There is no when-to-use guidance and no mention of alternatives such as upcoming_ai_model_shutdowns or list_ai_models. The agent must infer selection criteria entirely from the tool name.

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?

No annotations are provided, so the description carries the full burden. It does disclose one real behavioral constraint — the result set is capped at 10 models — which is useful, but it says nothing about what is returned per match, whether matching is fuzzy, or how auth/rate limits apply.

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, front-loaded sentence that leads with the action and then qualifies accepted inputs. No filler, though it packs three input modes and a limit into one dense clause.

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 annotations and no output schema, the description should explain what a match looks like (deprecation date, replacement suggestion, status). It discloses the 10-result cap but not the return shape or matching semantics, leaving a gap for a tool whose output drives a migration decision.

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 0% and the single 'text' parameter is undocumented in the schema, so the description must compensate — and it does, telling the agent the parameter accepts pasted code, config files, or a list of model IDs. That is meaningful beyond the bare 'string' type, though it could still clarify list formatting.

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 (find deprecated/about-to-be-deprecated models) and names the accepted inputs (code, config files, model ID list). It covers both the 'already shut down' and 'upcoming' cases, which makes it hard to distinguish from the sibling upcoming_ai_model_shutdowns, 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?

Usage is implied by the input types (paste code/config for a bulk scan), but there is no explicit when-to-use guidance or naming of alternatives such as upcoming_ai_model_shutdowns or find_ai_model_replacement. An agent can infer the scenario but is not routed to the right 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?

No annotations are provided, so the description carries the full behavioral burden, yet it discloses nothing beyond purpose: no indication that it is a read-only lookup, whether a model without a replacement returns empty or errors, or any auth/rate-limit considerations.

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 short, front-loaded sentence with zero wasted wording. Its brevity borders on under-specification rather than elegant conciseness, but structurally it is tight.

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?

This is a simple one-parameter read lookup with no output schema, so the description need not explain return values. However, with zero annotation coverage and an undocumented 'q' parameter, it leaves the agent guessing about input format and lookup failure behavior.

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 lone parameter is the opaque name 'q'. The description implies q is a model identifier ('某個模型'), which adds some meaning, but gives no expected format (name vs. ID vs. version) so the agent cannot construct a valid call confidently.

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: look up the officially recommended replacement model for a given model, which is a distinct operation from the sibling shutdown/list tools. It does not explicitly name any sibling to differentiate against, so it stops short of 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?

There is no explicit when-to-use or when-not-to-use guidance, and no alternative is named. The migration/replacement context is only implied by the word '改用', leaving the agent to infer that this is for deprecated models.

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?

No annotations are provided, so the description carries the burden. It usefully discloses what information is returned (status, dates, replacement, source), but never states that this is a read-only lookup or describes error behavior for unknown IDs.

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 front-loaded sentence that packs the resource, examples, and returned fields with no filler. Dense but not padded.

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 simple one-parameter lookup this is largely sufficient — the returned fields are enumerated even without an output schema. The main gap is the absence of routing guidance against the five sibling tools, which all touch the same model data.

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 only one parameter and 0% schema description coverage, the description compensates well by naming it ('模型 ID') and giving three concrete format examples (gpt-4o-2024-05-13, claude-3-5-haiku-20241022, gemini-2.0-flash) that show the expected identifier shape.

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 (查單一模型 ID) and enumerates the fields returned: status, retirement date, shutdown date, days remaining, official replacement, source. This clearly distinguishes it from the plural list_ai_models, though it never explicitly names a 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?

Usage is implied by '單一模型 ID' (one model, not a list or calendar), which contrasts implicitly with list_ai_models and ai_model_shutdown_calendar. However there is no explicit statement of when to use this versus find_ai_model_replacement or upcoming_ai_model_shutdowns.

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?

No annotations are supplied, so the description carries the full disclosure burden. It does add real behavioral content — it enumerates the returned attributes (淘汰狀態、下架日、官方建議替代) — and the verb 列出 implies a read-only listing. But it omits pagination, sorting, result volume, and data freshness, which matter for a catalog spanning three providers.

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 that front-loads what is returned and appends the filter capability. Zero filler and no repetition of the tool name or obvious facts.

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 three optional filter params, no output schema, and no annotations, the description covers the core (resource, returned fields, filters) but leaves the sibling-boundary question unresolved — with five related tools in scope, the agent still lacks a criterion for choosing this one. Adequate but with a clear 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%, so the description must compensate, and it does: '可依公司、狀態、類型篩選' maps directly onto provider/status/kind and confirms all three act as optional filters. The enum values themselves are self-explanatory, so no deeper syntax detail is needed.

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 (列出) and resource (AI 模型與 API 功能的淘汰狀態、下架日、官方替代), which is far more informative than the bare name list_ai_models. However, it never distinguishes itself from siblings like ai_model_shutdown_calendar, upcoming_ai_model_shutdowns, or find_ai_model_replacement, whose scope overlaps heavily with the deprecation/shutdown/replacement content it claims to return.

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 that results '可依公司、狀態、類型篩選' but gives no when-to-use guidance, no prerequisites, and no pointer to alternatives. An agent cannot tell from this text whether to call list_ai_models, upcoming_ai_model_shutdowns, or find_ai_model_replacement for a given deprecation question.

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?

With no annotations supplied, the description carries the full burden. It discloses the default time window (90 days) and the sort order (by shutdown date), which is useful behavioral context, but says nothing about the safety profile, return format, pagination, or what happens with large ranges. Adequate but thin for a zero-annotation 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?

A single compact sentence with the resource front-loaded, followed by the default and sort order. No filler or 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?

For a simple, zero-required-parameter read tool with no output schema, the description covers the essential semantics (default window, ordering). It lacks only routing guidance versus the similar calendar sibling, which is a minor 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% and there is one parameter, 'days', which the schema does not document at all. The description compensates by revealing the default value (90 days) and that it defines the lookahead window, but does not state units or accepted range behaviour.

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: listing AI models that will be decommissioned, with a scope (within N days) and an ordering (by shutdown date). It is clear what the tool returns, though it does not explicitly distinguish itself from the close sibling ai_model_shutdown_calendar.

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?

There is no guidance on when to use this tool versus alternatives such as ai_model_shutdown_calendar or check_code_for_deprecated_models. The default of 90 days is mentioned, but no conditions, prerequisites, or exclusions are given.

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