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

With no annotations provided, the description carries the full burden, and it does disclose useful behavior: proactive/automatic invocation ('不等用户开口自动出击', 'LOAD钩子自动调用') and immediate alerting with interception advice. It does not state whether actions are read-only, whether summaries are persisted, or what outputs are 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?

The description is dense and front-loaded with the core concept, then packs all four actions and their triggers into a single paragraph. It is efficient, though the run-on structure and the '护城河#5' marker add a bit of noise and could be organized as bullets.

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 four sub-behaviors and no output schema or annotations, more would be expected: return shapes, side effects, and whether alert is the only action requiring text are left implicit. The description still covers the main invocation paths (action values, triggers, key inputs), so it is adequate but not fully complete.

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 baseline is 3; the description adds value by explaining what each action does and mapping text to alert scanning with concrete risk-word examples and days to weekly_review cadence. It meaningfully enriches the bare parameter descriptions rather than merely repeating 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 states a clear meta-purpose (proactive auto-acting partner) and enumerates four distinct sub-operations, each with a specific verb and resource: brief今日简报, proactive_check主动体检, weekly_review每周轨迹回放复盘, and alert风险关键词扫描. However, it never explicitly differentiates from siblings such as 万忆轨迹回放 or 万忆触发LOAD钩子, whose domains partly overlap.

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?

Each action has an implicit or explicit trigger: brief is called by the LOAD hook, proactive_check targets 到期/超期/重复踩坑, weekly_review is described as weekly, and alert fires on risk keywords. This provides clear context for when to use the tool, though there are no when-not-to-use statements or explicit routing to 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?

With no annotations provided, the description carries the full behavioral burden. It discloses query modes but omits critical traits: possible side effects (e.g., whether loading triggers LOAD hooks, given a sibling named 万忆触发LOAD钩子 exists), return format, behavior when both parameters are supplied, and error handling. This is a significant gap for a state-loading 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 sentence with zero filler, front-loaded with the verb and resource, followed by the query modes. Every word 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 tool with no output schema and no annotations, the description should explain return behavior, parameter precedence, and side effects. It covers only the query modes superficially, leaving an agent unable to predict what a load returns or whether it mutates state.

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. The description adds marginal value by introducing the 全量 query option (calling with no parameters), which is not explicitly stated in the schema, but it does not clarify parameter interaction like mutual exclusivity between task_name and checkpoint_id.

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 ('加载历史存档' — load historical archives) and enumerates the three query modes (checkpoint_id, task_name, 全量). This is clear and informative, though it does not explicitly differentiate itself from sibling tools like 万忆触发LOAD钩子 or 万忆轨迹回放.

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 is given on when to use this tool versus alternatives. The query modes imply use cases, but there are no exclusions, no named alternatives, and no mention of whether this tool should be preferred over 万忆触发LOAD钩子 for state restoration or 万忆轨迹回放 for replay scenarios.

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 disclosure burden and does it well: it reveals automatic triggers (auto-open on interception), deferred behaviors (auto-settle when due, auto-verdict determination), and cross-tool side effects (counter_won auto-sinks into the error notebook and reinforces confidence interception). It omits output/return behavior and error handling, but the core behavioral traits are disclosed.

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 core mechanism is front-loaded, and each subsequent clause conveys distinct information: trigger condition, manual actions, verdict values, and downstream effects. The opening self-reference ('护城河#2「反事实之镜」') is mild filler, and the single dense run-on sentence is harder for an agent to parse than structured sentences, but nothing is wasted.

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 13 parameters and 6 actions but no annotations and no output schema, the description covers the open→settle→verdict lifecycle and downstream effects well. However, it omits the 'get' action that appears in the schema's action enum, says nothing about return values or list output shapes for list_open/list_settled, and provides no error or edge-case guidance—gaps an agent cannot fill from structured 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?

Schema description coverage is 100%, so every parameter is already documented in structured form and the baseline applies. The description adds workflow-level semantics (risk_level drives auto-generation of counter_path and settlement days; settle auto-judges verdict from outcomes), but it largely restates what the schema already says rather than adding meaning beyond it.

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 concrete mechanism: opening parallel fact/counterfactual branches at decision points and auto-settling them for comparison. It enumerates the sub-actions (open/settle/list_open/list_settled/auto_check_due), the verdict outcomes (fact_won/counter_won/neutral), and the workflow linkage, which clearly distinguishes it from the sibling memory/experience/confidence 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 clear trigger conditions: automatic branch opening on every BLOCK/CAUTION interception, plus manual open at key decision points, and explicit purposes for list_open/list_settled/auto_check_due. It does not name sibling alternatives or state when not to use the tool, which would push this to 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?

With no annotations, the description carries the behavioral disclosure burden. It does reveal non-obvious behavior: reranking, explicit temporal decay, and automatic gap recording on weak recall. However, it does not explain the scope or permanence of the gap-recording side effect, nor return behavior.

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 with the core retrieval concept, but it is a dense run-on packed with version markers and parenthetical implementation details. Version numbers like v5.0-v5.4 add noise without helping an agent decide how to use the tool.

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 seven parameters, no output schema, and no annotations, the description is incomplete for correct invocation. It does not state what the return value looks like, when this tool should be preferred, or how it differs from sibling retrieval tools.

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 schema already documents all parameters. The description adds some context for the use_graph parameter by mentioning the knowledge-graph channel, but it does not materially clarify parameter formats, ranges, or relationships 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 identifies 万忆召回记忆 as a retrieval tool by naming '检索' and specifying a four-channel hybrid retrieval process. However, it does not explicitly distinguish itself from sibling tools like 万忆图谱搜索, which overlaps with the knowledge-graph channel mentioned here.

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 provides no guidance on when to use this tool versus alternatives such as 万忆图谱搜索 or 万忆知识空白. It describes internal mechanics but not the intended selection criteria 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?

With no annotations, the description carries the burden. It usefully enumerates sub-behaviors of deep consolidation and says the daily-log mode 'only writes' the daily log. However, it does not state whether consolidation mutates existing memories, whether it is reversible, or what side effects occur.

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 text is compact, but the dense parenthetical list and the '/ 仅写每日日志' slash make the structure ambiguous. '园艺师后台（KektorDB式）' is also unexplained filler. A bulleted or explicit mode-by-mode layout would be clearer.

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 no output schema and no annotations, the description should say what the tool returns, when to choose each mode, and whether consolidation has side effects on stored memories. It covers only the high-level operations, leaving an agent without enough context to confidently select it over sibling consolidation tools.

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 covers the one action parameter, but the description adds meaning beyond the property comment: it expands 'consolidate' into four concrete analysis steps and clarifies 'daily_log' as writing only the daily log. That is valuable enrichment.

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 concrete operations: deep consolidation with contradiction detection, redundancy detection, insight extraction, skill crystallization, and daily-log writing. This is not a tautology and is reasonably specific, though the slash between the two modes is ambiguous and no sibling tool is named 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?

There is no guidance on when to call this tool versus siblings such as 万忆睡眠巩固 or 万忆知识压缩. The action choices consolidate/daily_log are listed, but no contextual trigger, 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.

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

No annotations are provided, so the description carries the behavioral transparency burden. It discloses the graph construction mechanism (automatic edge creation on write, with edge types semantic-similar/same-category) and the traversal nature of the search. However, it does not clarify whether the operation is read-only, how paging or result shapes behave, or any rate-limit/performance implications of multi-hop traversal.

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 one dense, front-loaded sentence that immediately states the action, resource, and mechanism. The parenthetical about v5.3 auto-edge-creation adds useful graph semantics without bloating the text. No filler or redundant restatement of the tool name exists.

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 parameters, no output schema, and no annotations, the description covers the main operational aspects: entity-based query, multi-hop depth, semantic association discovery, and the underlying edge types. It does not describe return-value structure or explicitly address edge cases, but for a retrieval-oriented tool the provided context is largely sufficient. The main gap is the absence of any statement about safety or side effects, but the description's '检索' wording implies read-only 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 100%, so the schema already explains query, depth, and limit. The description adds the conceptual framing that the query is an entity and depth corresponds to hops, but this is largely implied rather than explicitly tied to parameters. It does not add meaningful parameter semantics 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?

The description states a specific verb and resource: it performs knowledge-graph expansion retrieval via multi-hop traversal from an entity node to discover semantically related memories. This clearly distinguishes it from siblings like 万忆召回记忆 or 万忆轨迹回放 because it emphasizes graph structure and multi-hop traversal. The scope is concrete and not a tautology of the tool name.

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 makes the usage context reasonably clear: an agent should use this tool when it has an entity and wants to explore related memories through graph edges. It neither explicitly names alternatives nor gives when-not-to-use guidance, but the traversal-oriented framing implies when it is appropriate relative to sibling tools. No misleading guidance is present.

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 discloses a useful behavioral trait: cross-session persistence for cold resume. However, it does not state whether saving overwrites existing progress, what authorization or side effects are involved, or what the response looks like.

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 front-loaded sentence that delivers the core purpose, the key behavioral benefit, and the saved fields in minimal space. No filler or redundant phrasing.

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 definition is adequate for basic invocation, especially given the complete schema. However, with no annotations and no output schema, it does not explain return values, overwrite semantics, or error conditions, which are useful for an agent handling persistence operations.

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 schema already fully documents all four parameters. The description essentially restates the parameter list in prose without adding constraints, default values, or interaction details, so it adds no significant semantic value beyond the baseline.

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?

Description uses specific verb '存档' and resource '任务进度', and explicitly enumerates what is saved: task name, phase, progress percentage, and full state. It also mentions the cross-session cold-resume capability, making the tool's purpose unambiguous and distinct from the sibling '万忆加载进度'.

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 phrase '支持跨会话冷续传' clearly conveys the intended use case: persist task progress so work can be resumed across sessions. It does not explicitly mention alternatives or exclusions, but the context is clear enough for an agent to select this tool over its load counterpart.

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, the description carries the full behavioral burden. It does disclose a meaningful side effect: '同时存入道级钉住记忆' (also stored into Dao-level pinned memory). However, it does not clarify merge versus replace behavior, reversibility, or any access constraints.

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?

One compact sentence front-loads the main action, groups examples in parentheses, and appends the side effect. No redundant or filler content.

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 100% schema coverage, the description is mostly adequate but leaves gaps: no explanation of whether the data merges with existing profile fields, what '道级钉住记忆' fully implies, or what result the agent should expect given there is no 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?

The schema already fully describes the sole parameter 'data' as '用户档案数据（任意键值对）'. The description adds practical value by enumerating likely key categories such as trading style, risk preference, writing style, and personal info, helping an agent populate the arbitrary object.

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: '更新用户偏好档案' (update user preference profile), with concrete examples of covered fields. The target resource is distinct from all sibling memory tools such as 万忆错题本, 万忆经验库, and 万忆召回记忆.

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 parenthetical examples ('交易风格/风险偏好/写作风格/个人信息等') imply when to use the tool, but there is no explicit guidance about when not to use it or which sibling tool to choose instead. Usage is inferred rather than 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?

With no annotations, the description carries the full burden of behavioral disclosure. It conveys that this is an update operation, but it does not specify whether existing data is overwritten or merged, whether a missing section is created, what effects occur on other sections, or what the operation returns.

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, front-loaded sentence with no filler. The key action '更新交易策略锚点' appears first, followed by a concise explanatory clause. It is appropriately sized for the tool's apparent scope.

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 presence of a nested 'data' object, required parameters, no output schema, and no annotations, the description is too brief to be fully actionable. It does not define what an anchor is, what data structure is expected, or the semantics of updating a section, leaving significant ambiguity for an agent.

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 both parameters already have basic descriptions. The tool description adds the context of section-based management but does not clarify the internal structure of the 'data' object or provide parameter-level details 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 states a specific action and resource: '更新交易策略锚点' (update trading strategy anchor) and adds that it manages core trading logic and parameters by section. This clearly separates it from the memory-oriented sibling tools by domain. However, the concept of '锚点' is left undefined and no explicit sibling differentiation is made.

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 is provided on when to use this tool versus alternatives, such as creating a new anchor, storing progress, or loading progress. The description merely implies the purpose without stating conditions, 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?

没有 annotations，描述承担了行为说明负担。它明确这是“查询/可观测性”操作，并交代了返回内容和数据来源（真实发生在文件系统上的进化），行为透明。虽未显式说“只读”，但“查询”和“可观测性”已足够暗示无副作用。

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?

单句描述，破折号前给出核心信息，破折号后补充观测性与数据真实性，没有冗余内容。语言简洁且关键信息前置，每个部分都服务于帮助代理理解工具。

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?

工具只有 1 个可选参数，无 output schema，复杂度低。描述覆盖了目的、输出维度和数据来源，足以支撑正确调用；缺少明确的使用条件，但未影响基本完整性。

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?

input schema 覆盖率为 100%，唯一参数 limit 已在 schema 中有“返回条数”的描述，因此描述无需额外解释参数。描述本身也未增加超出 schema 的参数语义，符合无补偿需求时的基线 3 分。

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?

描述以“最近进化记录”开头，明确资源与查询意图，并具体列出“谁变了、为什么变、置信度变化”等返回维度。这使其与记忆、存档、导入等兄弟工具形成清晰区分，代理无需打开 schema 即可判断用途。

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?

描述没有说明何时使用此工具、何时不该使用，也没有点名任何替代工具。“任何聊天框都能查到”仅暗示通用可访问性，未提供选择依据或排除条件，使用指引明显不足。

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 provided, the description carries the full behavioral burden and does disclose core effects: memory decay, importance boosting, similar merging, and low-value archiving. It does not clarify whether the decay or archiving is destructive, reversible, or requires user confirmation, which is material for a maintenance operation.

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 with no filler; the main action is front-loaded and the four component operations are listed clearly. Every part 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 no-parameter batch operation, the description covers the main effect set but leaves ambiguity about side effects, reversibility, and how results are surfaced. Since there are no annotations or output schema, a bit more operational detail would make it fully actionable.

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 schema already exhaustively covers invocation requirements, which warrants the baseline score of 4. The description adds useful context about what the no-argument operation actually does.

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 identifies a specific batch operation (sleep consolidation) and enumerates its four sub-behaviors: memory decay, importance enhancement, similar merging, and low-value archiving. It is not a tautology, though it does not explicitly contrast with sibling tools like 万忆知识压缩 or 万忆记忆自检.

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 phrase '定期运行' indicates that the tool is intended for periodic scheduled maintenance rather than immediate on-demand use. However, it does not provide explicit when-to-use or when-not-to-use guidance, nor does it name any alternative tool to prefer in other situations.

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 provided, the description carries the full burden and handles it well: it discloses that compression is full-scale, that summaries and metadata are retained, that original content is archived rather than deleted, and that storage is freed. It does not discuss reversibility or permissions, so it stops short of a 5.

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 em-dash-joined sentence packs action, scope, retained data, archival behavior, and the storage benefit with no filler. Key information is front-loaded and 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 simple one-parameter tool with no output schema, the description is largely complete: the target and outcome are clear, and the archival guarantee addresses the main safety concern. It lacks explicit usage guidance and any statement about reversibility, but those are modest gaps at this complexity level.

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 of 3 applies. The description only echoes 'specified layer' and adds no meaning beyond the schema's layer enum and default value, which are already documented in the input 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 states a specific verb and resource: full compression of a specified layer, with concrete outcomes (retain summaries/metadata, archive originals, free storage). It is clearly the compression tool among the siblings, though it does not explicitly name or contrast an alternative.

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 phrase '释放存储空间' implies the intended purpose and '指定层级' implies scope selection, so usage context is somewhat implied. However, the description gives no explicit when-to-use guidance, no exclusions, and no mention of alternative 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?

With no annotations, the description carries full burden; it discloses that the tool auto-records gaps when recall is weak, sorts list by hit count, requires gap_id+note for close, and explains the behavioral aim of confident-vs-honest responses. It stops short of describing persistence, auth, or side effects beyond recording gaps, but for a memory-management tool this is strong 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?

Four sentences, front-loaded with the metacognitive purpose, then dense but parseable operation specs. Every clause earns its place—no fluff. Technical version number and framing terminology are compact and meaningful for an advanced agent.

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 4 optional params, no output schema, and no annotations, this is nearly complete: it covers purpose, operations, parameter roles, and behavioral philosophy. It doesn't state what list/stats return structurally, but the description communicates intent well enough for selecting and invoking; remaining gap is minor.

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. The description adds meaning by tying each action to its purpose, clarifying that gap_id is for close and note is the supplement source, and that list sorts by hit frequency. This goes beyond raw schema descriptions meaningfully.

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 resource (knowledge gaps) and the three operations (list/close/stats) with clear verbs. It also distinguishes this tool from siblings like 万忆记忆自检 and 万忆召回记忆 by framing it as the 'knowing what you don't know' mechanism tied to retrieval weak spots.

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 explicitly states when to use the tool (when recall is too weak, when system should acknowledge weak inventory) and how it links to semantic vector retrieval. It doesn't explicitly say 'when not to use' or name alternatives, but the metacognitive framing and operation list imply usage boundaries clearly enough.

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 present, so the description must disclose behavior. It only describes the content/purpose in nominal terms; it does not say whether the call is read-only, whether it returns a list or a single pattern, or how limit/task_name affect execution. '沉淀' even introduces ambiguity about whether the tool stores or retrieves.

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 text is extremely short and contains no filler, but it is under-specified rather than feature-complete. A single noun-phrase clause conveys purpose without separating what the tool does from when it should be used.

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 the return shape and invocation behavior; it instead only restates the conceptual role of the experience library. The optional parameters are documented in the schema, but the high-level behavior and output remain unspecified.

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%: limit is documented as 返回条数 and task_name as 按任务筛选. The description adds no parameter detail, but the schema already carries the full burden, so a 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 identifies the resource (经验库) and its content (成功路径沉淀的可复用模式), and hints at cross-task reuse, so it is not a tautology. However, it lacks an explicit action verb such as '查询' or '检索', leaving the agent to infer that this tool retrieves experience patterns rather than, say, creating or consolidating them.

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?

'跨任务迁移复用' offers a plausible use case: retrieve reusable patterns from successful paths to apply to new tasks. But the description does not explicitly state when to prefer this tool over siblings like 万忆轨迹回放 or 万忆错题本, nor does it mention 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?

With no annotations provided, the description carries the full burden. It discloses that the tool can BLOCK low-confidence decisions, surface historical mistakes, and automatically trigger moat #2 to open counterfactual branches. This reveals critical side effects and control-flow behavior beyond the bare 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?

The description is dense and front-loaded with the core purpose, then flows into action semantics, usage instructions, and behavioral consequences. It is slightly long and packed into a single paragraph, but every sentence contributes useful information with minimal waste.

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 tool has 8 parameters, no annotations, and no output schema, the description covers the main decision points: supported actions, required inputs, high-risk trigger contexts, and block/linkage behavior. It does not describe return values or error conditions, but the essential selection and invocation guidance is present.

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 baseline is 3. The description adds meaning by explaining the role of each action value, indicating which actions require decision_text, and clarifying validate/challenge as support/opposition votes. This goes beyond the schema's short field descriptions.

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 defines the tool as a cognitive confidence check for high-risk decisions, enumerates all seven supported actions (check/validate/challenge/review/get/rank/needs_review), and names concrete risk scenarios. This makes the purpose specific and distinct from the sibling memory 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?

It explicitly states when to use the tool: '在做重仓、追涨、删库、强推等高风险动作前必调用' (must call before high-risk actions). It does not discuss when not to use it or name a specific alternative, but the mandated usage context is strong and actionable.

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. It does disclose the main behavior: automatic memory injection at startup, including pinned/high-frequency memory and checkpoint-resume checking. However, it omits side effects, failure handling, read-only vs. mutating behavior, and what happens when no memory exists.

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, front-loaded sentence. Every clause adds useful information about the hook type, the trigger, the action, and the memory categories involved, with 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 parameter-free hook, the trigger and primary behavior are present, and there is no output schema to account for. However, the internal terminology is undefined, and the lack of side-effect or invocation details leaves the description adequate but not fully complete for an agent deciding whether or how to call it.

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 and schema description coverage is 100%, so the baseline is 4. There is no parameter detail for the description to add, and it correctly avoids inventing any.

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 identifies the operation ('auto-inject memory'), the trigger ('session startup'), and the memory categories involved ('pinned', 'high-frequency', 'checkpoint-resume check'). It is distinct from the STORE hook and load-progress siblings, though it does not explicitly name an alternative sibling and relies on unexplained internal terms like '道级' and '法级'.

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 trigger context is stated explicitly: this hook runs automatically at session startup. However, there is no guidance on when not to use it, whether it should ever be manually invoked, or how it compares with related siblings such as 万忆加载进度 or 万忆触发STORE钩子.

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, the description must carry behavioral disclosure; it explains that the tool archives and reflects, extracting decisions, patterns, skills, and postmortems. However, it does not disclose side effects such as persistence, overwriting, or whether reflection results are stored separately, leaving some behavioral ambiguity.

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 hook type and core action. It is not bloated, though the 'HOOK-STORE / HOOK-REFLECT' prefix partly repeats 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?

Given 8 optional parameters, no annotations, and no output schema, the description is too sparse. It does not explain the mode parameter, what a successful call returns, prerequisites, or how this hook relates to the sibling LOAD hook and other memory tools.

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. The description's mention of decisions, patterns, skills, and postmortems mirrors existing parameter names but adds no meaning 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?

The description states a specific action ('归档 + 反思') and identifies the target artifacts ('决策、模式、技能、复盘'), making the tool's purpose clear. It is distinguishable from the LOAD hook by the STORE/REFLECT framing, though it does not explicitly name any sibling or contrast.

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 scopes usage to '会话结束时' (at session end), which is a clear usage context. It does not mention when not to use it or compare it with alternatives like LOAD hooks, but the timing guidance is practical and sufficient.

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 useful automatic behavior: classification, tagging, and indexing. But it does not explain output behavior, side effects on existing memories, privacy handling, or whether writes are idempotent.

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 with no filler. It conveys the core action and the key automatic behaviors efficiently.

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 12 parameters, no annotations, no output schema, and 22 sibling tools, the description is too minimal to fully orient an agent. It does not guide selection among sibling tools, clarify the relationship between automatic tagging and user-provided tags, or describe what happens after recording.

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. The tool description adds no parameter-level meaning, and its claim of automatic tagging could theoretically conflict with user-provided tags/category, which is left unexplained.

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 action (record) and a specific resource (full memory library), and adds automatic classification, tagging, and indexing behavior. It is clear what the tool does, but it does not explicitly distinguish this from specialized sibling tools such as 万忆经验库 or 万忆错题本.

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 implies the tool should be used for new knowledge, observations, or experiences via '将一条新知识/观察/经验记录入'. However, there is no explicit guidance on when not to use it or how it differs from sibling memory-recording/import 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?

With no annotations provided, the description carries the full burden and does disclose the behavioral scope by listing the health dimensions it evaluates. The term '检测' implies a read-only diagnostic operation, and the enumerated status checks give the agent a concrete sense of what will be inspected, though side effects and output format are 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?

The description is a single compact line that front-loads the core purpose and then compactly lists the specific checks. There is no redundant or filler text.

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 zero-parameter diagnostic tool, the description covers the essential information: what is being checked and at what level (full-library). It does not specify output format or performance cost, but given the simplicity of the tool this 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?

The tool has zero parameters, so there is no parameter meaning to convey. The description appropriately focuses on the tool's checks instead of parameter details.

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 ('检测') and resource ('全库健康度') and enumerates the exact aspects checked: total memory count, layer distribution, index status, FTS status, and graph status. This clearly distinguishes it from sibling tools focused on operations like recording, recalling, or searching.

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 implies the tool is for diagnosing overall memory health, but it does not explicitly state when to prefer it over alternatives or mention any prerequisites or exclusions. The usage context is inferable from the content but not directly guided.

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?

没有annotations，描述承担了主要行为披露责任；它说明了跨域桥接、BLOCK/CAUTION自动联动、按命中率列表、查看详情等行为，比完全沉默好。但未说明abstract沉淀模式是否有持久化副作用、bridge返回什么结构、get和list失败时的行为，行为透明度仍不完整。

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?

两句话覆盖了四个子操作和核心跨域迁移逻辑，信息密度高，动作列举紧凑，没有明显冗余。缺点是「护城河#3」和未解释的BLOCK/CAUTION术语占用了一定理解成本，但整体仍算精简。

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?

在无output schema、无annotations的情况下，描述覆盖了四个action的主要用途和参数约束，并暗示了部分返回行为（如list_patterns按命中率、get看详情）。但缺少对输出形状、limit如何应用、domain如何筛选、以及抽象模式后是否返回ID或确认信息等说明，对一个9参数的多模式工具来说只是刚好够用。

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?

input-schema对所有9个参数都有描述，且已经标注了abstract必填、bridge必填、get必填等条件，因此description无需重复基础语义。描述额外将参数组合归入四个action，便于快速理解调用姿势，但未补充limit作用范围、domain如何影响bridge结果等更深层语义，价值有限。

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?

描述明确说明工具的核心动作：把交易/写作/开发各领域的教训抽象成跨域底层模式，再进行跨域桥接，并能与兄弟工具如「万忆经验库」「万忆图谱搜索」区分开。但开头「护城河#3」和使用BLOCK/CAUTION这类内部术语，使整体表达不够口语化，且未直接对比任一兄弟工具，因此未到5分。

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?

描述按action列举了abstract、bridge、list_patterns、get各自的用途与参数组合，并提到BLOCK/CAUTION拦截时自动联动，提供了可执行的调用语境。但没有明确说明何时应优先使用本工具而非其他记忆工具，也没有给出when-not或替代工具的显式指引。

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, and it does so well by detailing what each action does: timeline replays all branches chronologically, stats provides specific metrics like settlement rate and risk distribution, route compares actual versus all-obey paths, and review summarizes recent decisions. It does not explicitly state read-only behavior or output structure, but the visualization/replay framing makes the non-mutating nature reasonably clear.

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 well structured: core purpose first, action breakdown second, and user benefit last. The '护城河#4' branding and the somewhat repetitive closing sentence add minor noise, but overall every section earns its place and the length is appropriate.

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 three optional parameters and no annotations or output schema, this description covers all significant behavior: the four modes, the main data categories, and the role of the review window. It does not describe exact response shapes or edge cases, but it is sufficient for an agent to select the tool and invoke it with reasonable confidence.

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 baseline is 3, but the description adds real value by explaining the semantics of the action values and linking days to the review window and limit to timeline. This helps an agent understand how the parameters affect output beyond their terse schema descriptions.

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's core purpose: replaying historical counterfactual branches into a decision timeline and visualizing decision evolution. It enumerates four concrete action modes (timeline/stats/route/review), giving a specific verb and resource. However, it does not explicitly differentiate itself from potentially overlapping siblings like 万忆反事实之镜 or 万忆查询进化.

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 implies that the tool is for retrospective decision-career visualization and analysis, with each action serving a distinct purpose. It does not explicitly say when to prefer this tool over alternatives or when not to use it, leaving the selection largely 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?

With no annotations provided, this description carries the burden, and it does disclose important side effects: failures automatically become mistake-book entries, successful conclusions automatically become experience-library entries, and an anchor_state enables breakpoint recovery. It does not mention idempotency, overwrite behavior, or authorization requirements, but the core behavioral profile is transparent.

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 one information-dense sentence with every clause doing work: scope, phase values, automatic side effects, and anchor capability. It could be lightly restructured for readability, but it contains no filler and leads with the core purpose.

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 6 parameters and a nested anchor_state object, the description covers the key behavioral contracts: what gets stored, how outcomes are classified, and why anchor_state exists. It does not explain what the call returns, whether process_id creates vs continues a trace, or the exact structure of anchor_state, but the combination with the detailed schema is sufficient for correct invocation.

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 baseline is 3, but the description adds useful semantics beyond the schema: it enumerates the five phase values, explains what outcome=failure does automatically, and gives purpose to anchor_state (断点恢复). This extra context helps the agent choose correct values even though the schema already describes 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?

The description states the exact operation — store one phase (planning/attempt/error-correction/reflection/conclusion) of a task execution into a trajectory archive — and distinguishes this from sibling tools by noting automatic precipitation into the mistake book and experience library. It names the resource (轨迹档案) and the scope (task execution phases), so an agent can identify 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?

The description makes clear it is for recording intermediate execution phases and that failures/successful conclusions are auto-routed to 万忆错题本 and 万忆经验库, which tells the agent when this tool is the right choice. It does not explicitly state exclusions or name an alternative for querying/replaying traces, so it misses the top bar for explicit when-not 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?

当前没有任何注解，因此描述承担了行为披露的全部责任。它披露了'失败决策自动沉淀为反例条目'和'跨会话'这两点，说明工具会自动记录失败且数据具有持久性，这对agent理解副作用有实际价值。但描述未提及补写教训是否产生持久写入、是否需要权限限制，也没有说明返回形式，信息仍不够完整。

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?

描述只有两句话，第一句列出核心功能，第二句补充关键行为，信息前置且没有冗余词。每个分句都贡献了独立信息，括号中的'Reflexion式'也提供了方法论线索，整体紧凑高效。

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?

描述给出了明确的目的和一个关键行为，参数层面也由schema完全覆盖，但缺少输出schema和注解，导致agent无法预知调用返回的内容格式，也没有失败或边界场景的提示。对一个4个可选参数的工具来说，描述达到最低可用标准，但仍有明显信息缺口。

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?

输入schema对所有4个参数都有详细描述，schema_description_coverage为100%，所以参数含义主要由schema承担。描述中的'查看/补充'只是重复了action枚举（list/patterns/lesson）的含义，没有为参数添加新的语义或边界条件，因此按基线给3分。

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?

描述以'错题本（Reflexion式）：查看错题/高频错误模式/补充教训'开头，明确列出了具体动词（查看、补充）和资源（错题、模式、教训），并借助'失败决策自动沉淀为反例条目'这一特性与其他记忆类兄弟工具产生区分。但没有显式点名与哪个工具不同或给出对比条件，因此未达到满分。

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?

描述用'跨会话防重复犯错'明确给出了使用情境——在会话间回顾错误教训以避免重蹈覆辙，这为agent提供了清晰的调用上下文。但它没有说明何时不使用、也没有列出替代工具或排除条件，所以给4分而不是5分。

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