Server Quality Checklist

Tool Scores

• ask_webA
  4.1/5.0
  Behavior4/5

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

  With no annotations, the description carries the full burden and reveals meaningful behavior: an internal sub-agent performs iterative searches, opens web pages, cross-checks information, and returns a concise conclusion with sources. It also warns that the operation consumes a lot of context. It does not mention failure modes or external service dependencies, but covers the main behavioral traits.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  Two dense sentences: the first defines what it does, the second when to use it and what behavior to expect. Every clause contributes value and there is no redundant filler.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness5/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  The tool has only one parameter, no output schema, and no annotations. The description explains the operation, the internal process, the output format (conclusion with sources), and the appropriate scenarios, making it complete for an agent to select and invoke the tool correctly.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters3/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema description coverage is 100%, so the single 'question' parameter is already well documented. The description adds context about the kind of questions to ask and notes the research will consume context, but it does not materially extend the parameter's meaning beyond the schema.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose4/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  The description clearly states the tool performs web research ('联网查资料') and describes the multi-step behavior: searching, opening pages, cross-verifying, and returning a concise conclusion with sources. It differentiates from siblings implicitly via '联网', but does not explicitly name ask_memory or delegate, so it loses the top point.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines4/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  It explicitly identifies suitable use cases: checking a library version, a fact, or a price, and characterizes these as context-consuming research tasks. This gives clear usage guidance, though it does not name alternatives or state when NOT to use the tool.

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

• ask_memoryA
  4.2/5.0
  Behavior4/5

  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 explicitly discloses that the tool returns a retrieved/synthesized answer and states that it will honestly report when the memory store is empty or no match is found. This is strong behavioral coverage for a simple retrieval tool, though it does not explicitly confirm that it makes no modifications.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  The description is three short sentences, each earning its place: define the resource, state the use case, and describe return behavior plus edge cases. There is no filler, no repetition of schema content, and the most important distinguishing information is front-loaded.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness4/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  For a one-parameter tool with no output schema and no annotations, the description covers the resource, the intended use case, the return behavior, and the empty/no-match edge case. It does not explicitly discuss side effects or contrast with sibling tools, but nothing essential appears missing for correct invocation.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters3/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema description coverage is 100%, and the single parameter 'question' already includes an explanatory example ('想回忆的内容，如...'). The tool description reinforces the memory-store context but adds no materially new parameter-level details, so the baseline score of 3 is appropriate.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  The description opens with '问自己的长期记忆库（本地 markdown 文件）', giving a specific verb, resource, and storage format. It lists the content domains it covers (preferences, facts, notes, conclusions) and is implicitly distinguished from the sibling ask_web by emphasizing '自己的长期记忆库'. This leaves no ambiguity about what the tool does.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines4/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  The phrase '用于回忆此前记住的偏好、事实、笔记、结论' gives clear guidance on when to use this tool. It does not explicitly state when not to use it or name alternatives like ask_web or delegate, so it falls short of a 5, but the intended use case is unmistakable.

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

• delegateA
  4.4/5.0
  Behavior4/5

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

  With no annotations provided, the description carries the full behavioral disclosure burden. It reveals that the sub-agent can search the web, scrape pages, read/write memory, operate over multiple turns, and return a final result. It also warns that the task must be self-contained because the sub-agent cannot see the main conversation. It stops short of mentioning cost, latency, or side-effect details, but the core behavior is well disclosed.

  Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

  Conciseness5/5

  Is the description appropriately sized, front-loaded, and free of redundancy?

  The description is three short sentences with no filler. It front-loads the main action, then gives capability context, examples, and a critical invocation warning. Every sentence earns its place.

  Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

  Completeness4/5

  Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

  For a tool with no annotations and no output schema, the description provides enough for correct selection and invocation: what it does, what the sub-agent can do, typical use cases, and the self-contained task requirement. It could be more complete by explicitly routing simple queries to ask_memory/ask_web and noting possible side effects of memory writes, but overall it is sufficient.

  Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

  Parameters4/5

  Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

  Schema coverage is 100%, so the schema already documents task and background. The description adds meaningful semantic guidance: the task parameter should describe a multi-step chore, be self-contained, and target research/organization/memory-writing goals. This goes beyond simply restating the schema.

  Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

  Purpose5/5

  Does the description clearly state what the tool does and how it differs from similar tools?

  Description states a specific verb and resource: delegate an entire multi-step chore to a sub-agent. It clearly distinguishes this from the sibling tools ask_memory and ask_web by emphasizing multi-step autonomous execution rather than a single direct query.

  Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

  Usage Guidelines4/5

  Does the description explain when to use this tool, when not to, or what alternatives exist?

  The description gives concrete suitable use cases: research summarization, organizing materials, and storing conclusions into memory. It implies this is for whole tasks that need multiple steps, but it does not explicitly state when not to use it or mention the sibling tools by name as alternatives.

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