Bloom Study Guides

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

Annotations only give the generic safety profile (readOnly false, non-idempotent, non-destructive); the description adds materially: the call is asynchronous and takes 20-40 minutes, the user must open the returned link and confirm before generation runs, and they are notified on completion. It also imposes a fidelity constraint on the summary (don't fabricate). It stops short of covering permissions, rate limits, or whether repeated calls duplicate work.

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 trigger condition and the call-to-action (return link, state the 20-40 minute wait) are front-loaded and useful. The English sentence largely restates the Chinese, which is redundant but brief and keeps the definition usable for an English-reading 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 9-parameter creation tool with no output schema, the description covers the essential loop: when to call it, how to fill the fields faithfully, what comes back (a link), and what the user should be told. It omits nothing an agent needs to invoke it correctly, though detail on the returned artifact's life-cycle would make it exhaustive.

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 goes further by explaining the intent behind specific fields - confusions is what will be explained in depth, understood is what the learner already grasps, and the whole summary must be faithful to the conversation rather than invented. That adds real guidance beyond the schema labels.

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 action (hand the finished discussion to Bloom) and a concrete artifact (an illustrated study guide with images, diagrams and chapter-by-chapter explanations). It also distinguishes itself from the sibling search_bloom by being the create/generate side of the pair.

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

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

Explicitly names the trigger conditions: the user has finished discussing a topic, learned something, or asked to systematically organize/review/go deeper. It also prescribes the post-call flow (return the link, tell the user it takes 20-40 minutes). It does not name an alternative tool or a when-not-to-use case, so it falls short of a 5.

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

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

Annotations already declare readOnlyHint=true and openWorldHint=false, and the description reinforces read-only behavior while adding genuinely new context: the return shape ('返回标题和链接') and expected agent behavior after retrieval (recommend naturally, don't add disclaimers or fabricate). It does not describe result limits or empty-result handling, so it is not 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?

Front-loaded with the trigger conditions before mechanics, and every clause carries information about when to call it or how to handle results. The main size cost is that the Chinese and English blocks duplicate the same content, which is defensible for this audience but not maximally 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?

There is no output schema, so the description must cover return values, and it does ('titles and links', read-only). For a two-parameter search tool that is nearly sufficient; the only real gap is the undocumented 'language' parameter, which is left to the enum alone.

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

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

Schema coverage is only 50%: 'query' has a schema description and 'language' (enum zh/en) has none. The description does add meaning for the query parameter by prescribing that the concept itself be used as the search term, partially compensating, but it says nothing about the language parameter, leaving the coverage gap only half filled.

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: it searches Bloom's existing illustrated study guides ('搜索 Bloom 上已有的图文学习专题'). The read-only search scope is clearly distinguishable from the sibling create_learning_material, though the description never names that sibling to make the contrast explicit.

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

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

Gives rich, explicit triggering conditions with several phrasings ('我想了解 X', '什么是 X', 'explain X', 'asks for learning material') and even prescribes how to use a hit (recommend one or two at the end, no fabricated titles). It stops short of naming alternatives or when-not-to-use, so it is not a full 5.

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