weReci

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

Beyond the annotations (readOnly=false, idempotent=true, destructive=false), it discloses that re-adding a recipe replaces its lines, that recipe_ids must come from specific tool results and never be invented, and that share controls whether lines land on the shared or phone list with a joining side effect. These are meaningful behavioral details that the annotations alone do not provide.

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 but every sentence carries operational guidance: call conditions, parameter usage, alternatives, idempotency, and provenance. The core action and the most important guardrail are front-loaded before details.

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 4-parameter mutation tool with no output schema, the description covers selection, invocation, side effects, and the two key sibling behaviors. It also provides explicit no-go cases and exact recipe-id provenance, so an agent has what it needs to invoke safely and correctly.

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

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

Even though schema descriptions cover 100% of parameters, the description adds cross-parameter constraints such as "at least one of the two", explains that recipe ingredient lines go onto the list, and instructs the agent to pass exclusions and "mention what you skipped". It also adds provenance requirements for recipe_ids not present in 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 opens with a specific action and resource: "Put things ON the user's own shopping list (List It)", and it clarifies the collection-only role. It also distinguishes itself from build_shopping_list by saying that consolidated/hand-off flows belong there, so an agent can tell them apart without opening schemas.

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

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

It gives explicit call conditions: only when the user asks to "add, save, or collect for later", and explicit prohibitions: "never speculatively, and never as a follow-up they didn't ask for". It names build_shopping_list as the alternative for proposing or handing off a list, and sets the at-least-one-parameter and share-omission rules.

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 provide only readOnlyHint=false, idempotentHint=false, and destructiveHint=false, which is minimal. The description compensates fully by disclosing that nothing is stored, that specific permissions (grocery:cart, email:send) are required, that the first use of cart mode may prompt account connection, and that coupons mode can only match items on coupon so most items remaining unmatched is normal. This exceeds what annotations offer and gives the agent a complete picture of side effects and conditions.

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 long but every sentence carries essential operational information. It is front-loaded with the core action and the propose-first rule. While it could be formatted with bullet points for easier scanning, the density is appropriate given the complexity of the tool's modes and permissions. There is no redundancy or fluff, so it earns a high score, though slightly below perfect due to the dense single-paragraph structure.

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 description covers all aspects needed for correct usage: the action, the sequencing requirement, all send modes and their permissions, the exclusion parameter semantics, the recipe_ids provenance rule, and the fact that nothing is stored. It even anticipates common pitfalls like unmatched items in coupons mode. For a tool with this complexity and no output schema, the description is remarkably 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?

Although the schema already provides detailed descriptions for all parameters (100% coverage), the tool description adds significant meaning beyond the schema: it explains the propose-first workflow that dictates how the 'mode' parameter must be used, clarifies the coupon matching limitation for the 'coupons' mode, and reinforces that recipe_ids must come from specific tools. This contextual guidance is not present in the schema and is essential for correct invocation.

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 purpose: composing a consolidated grocery list from 1-8 recipes and handing it off for shopping. It distinguishes itself from sibling tools like get_shopping_list or update_shopping_list by emphasizing the build-and-handoff nature. The verb 'compose' and resource 'shopping list' are specific and unambiguous.

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 explicit usage rules: always call with mode 'propose' first, show the user, then wait for explicit approval before using send modes. It lists each send mode with its conditions and permissions, and explicitly says 'NEVER propose and send in the same turn'. It also instructs that recipe_ids must be copied exactly from search_recipes or list_conceits results. This is comprehensive and leaves no ambiguity about when and how 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.

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

Annotations declare readOnlyHint=true, and the description aligns by saying it 'Returns the labeled path or reports they're unconnected' — a read-only explanation operation. It adds useful behavioral context: it returns the shortest chain, reports unconnected cases, and explains how user-supplied metadata nodes behave. It doesn't mention performance or limits, but for a read-only graph query the description covers the key 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 dense but well-organized: it starts with the core purpose, then gives the node-id format, then explains special node types. Every sentence adds useful information. It's longer than average, but the complexity of the node-id system justifies the length. Could be slightly tightened, but no wasted sentences.

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 read-only graph query with no output schema, the description covers the essential context: what the tool does, how to format node ids, how special metadata nodes work, and what the return is (labeled path or unconnected report). It doesn't explain edge_types semantics in detail, but the enum values are self-explanatory and the schema documents them. The description is complete enough for an agent to call this tool correctly.

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

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

Schema coverage is 100%, so the schema already documents all three parameters. The description adds significant value by explaining the node-id format ('<type>:<key>') with concrete examples and how bare recipe ids auto-resolve. It also explains the semantics of special node types like 'tag:', 'property:', and 'component:' — meaning beyond the schema's terse 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 opens with a specific verb ('Explain HOW') and a precise resource ('the shortest chain of relationships between any two nodes'), then gives concrete examples of node pairs. It clearly distinguishes itself from sibling tools like traverse_graph by focusing on shortest-path explanation rather than general traversal.

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 rich context on when to use the tool: to explain relationships between any two archive entities, with examples like 'what else uses this sauce?' It doesn't explicitly name alternatives or exclusions, but the node-id format guidance and relationship examples make the intended use clear. It could be stronger by explicitly contrasting with traverse_graph.

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

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

Beyond the readOnlyHint annotation, the description discloses important behavior: a job whose invocation died is reported as 'failed' rather than remaining 'running' forever. This tells the agent that polling will eventually terminate, which is valuable behavioral context not available from annotations alone.

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 exceptionally concise: two sentences that state the tool's purpose, return values, and failure semantics. Every sentence adds essential information, and the most important usage detail is front-loaded.

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 polling tool with readOnlyHint and no output schema, this description covers everything needed: what it checks, what states are returned, and how failures are handled. No critical information is missing.

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

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

The schema already fully describes the single required parameter, job_id, as 'The job id from start_background_job.' The description reinforces this by referring to jobs started with start_background_job, but does not add significant new meaning beyond the schema. Baseline 3 is appropriate given 100% schema coverage.

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 purpose: checking a background job started with start_background_job. It specifies the resource (background job) and the action (check), and defines the possible return states. This differentiates it from sibling tools like start_background_job and get_scale_job.

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 clear usage context: it is meant for checking jobs created by start_background_job and is safe to poll until the status stops being 'running'. It implies the polling pattern without explicitly listing alternatives or exclusions, but the guidance is sufficient for correct use.

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

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

Beyond the readOnlyHint annotation, the description discloses that results are exact and complete, returns are grouped by relationship, node IDs follow a type:key convention, bare search_recipes IDs are auto-prefixed with 'entry:', and metadata/component nodes are traversable. This is substantial non-obvious behavior that an agent needs to know.

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 long but every sentence carries necessary information, starting with purpose and exactness, then id syntax, then special node types. Examples are used instead of prose, and there is no filler or tautology.

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 description fully covers how to identify and invoke against a node, including unusual id types and auto-prefixing. Without an output schema, however, the exact response shape beyond 'grouped by relationship' is left somewhat implicit; pagination or limit behavior is also not mentioned. 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?

Although the schema already documents the single 'id' parameter at 100% coverage, the description adds concrete id formats and examples for entry, ingredient, author, appliance, user, tag, property, and component nodes. This goes well beyond the schema's one-line description and removes ambiguity about how to construct the parameter.

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 precise action and resource: return everything directly connected to a single graph node, grouped by relationship type. The one-node scope and 'Exact and complete (unlike semantic search)' line separate it from siblings such as traverse_graph, find_graph_connection, and get_graph_overview without needing to open their schemas.

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 intended use clear: when you need exact, complete one-hop graph relationships, and contrasts this with semantic search. It does not explicitly name alternatives like traverse_graph or provide when-not-to-use rules, but the examples ('what else uses this sauce?') give concrete invocation context.

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?

The readOnlyHint annotation already signals a safe read operation. The description adds behavioral detail beyond that by explaining the return shape: title, ingredients, instructions, notes, photo URL, and entry_kind for non-recipe entries. This is useful context that the annotation alone does not provide.

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

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

Two sentences deliver the core purpose, id provenance, output fields, and mixed-type behavior without redundancy. The most important 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?

With only one parameter and no output schema, the description carries the responsibility of explaining both input and output. It fully covers the input source, the fields returned, and how non-recipe entries behave. No critical information is missing for an agent to call this tool correctly.

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

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

The schema already documents the single id parameter with 100% coverage, so the baseline is 3. The description adds value by explaining where ids come from (search_recipes or list_conceits) and confirming that the id refers to a general entry, not only a recipe. This provenance helps the agent supply the correct value.

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 a specific action (fetch full details) and resource (one entry by id), and explicitly clarifies that this applies to both recipes and non-recipe entries. It differentiates itself from search_recipes and list_conceits by saying ids come from those tools, so an agent knows exactly what this tool is for.

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

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

The description explains that ids come from search_recipes or list_conceits results, which gives clear context on when to invoke this tool. It doesn't explicitly list exclusion criteria or name alternatives, but the source-of-id guidance is enough to route the agent correctly in most cases.

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

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

Beyond the readOnlyHint annotation, the description surfaces meaningful behavior: the list is the shared List It list, not the phone-only one; it returns line-level detail; it includes an `available` flag and a `note`; and it explains the `since`/`seq` mechanism for a cheap `unchanged` response. This is rich context an agent needs to handle the tool correctly.

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 information-dense and front-loaded: scope first, then usage, then a key caveat, then parameter guidance. Each sentence earns its place. It is longer than a minimal description, but the extra context (phone-only list, note, unchanged behavior) is essential for correct calling.

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 read-only tool with one optional parameter and no output schema, the description covers the returned data (lines, crossed-off states, aisles, recipe sources), the unavailable case (`available: false`, `note`, don't retry), and the incremental read via `since`. Nothing essential is missing: an agent can call it correctly and interpret the response.

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 documents `since` as 'The `seq` from your previous read. Omit for a full read' (100% coverage). The description adds an important behavioral nuance: passing `since` yields a cheap `unchanged` answer when nothing moved. This clarifies the parameter's practical purpose beyond the raw 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 starts with 'Read the user's SYNCED shopping list', giving a specific verb and resource, then enumerates the contents (lines, crossed-off state, aisle, recipe sources). It explicitly distinguishes the shared synced list from the phone-only list, and the use of 'SYNCED' plus the sibling tool names (add/update/build) makes its scope unambiguous.

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

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

It gives concrete triggers: 'Use it to answer "what's on the list?" or before ticking things off.' It also tells the agent not to retry when `available` is false and notes the phone-only list is not visible. However, it never names a sibling alternative (e.g., add_to_shopping_list for writes), so the 'vs alternatives' part is only implicit.

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?

The annotations declare readOnlyHint=true, and the description adds meaningful behavioral detail beyond that: it explains member kinds, the `reference: true` flag for Wikipedia ingredient cards, the `app_url_template` substitution rule, and that `unavailable_count` reports unresolved members. This gives the agent a solid understanding of what the result contains without contradicting the annotations.

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 yet information-dense. Every clause adds value: the definition of curated shelves, the member structure, the reference flag, the app link template, and the unavailable_count meaning. It is front-loaded with the action and resource, and there is 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?

Even though there is no output schema, the description supplies all essential return-value semantics: the top-level fields, the member object shape, the allowed kinds, the reference flag, the link-building instruction, and the meaning of unavailable_count. With zero parameters and a read-only annotation, this is sufficient for an agent to invoke the tool and interpret its result correctly.

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

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

The tool has zero parameters, so there is nothing for the description to clarify about argument semantics. The schema already covers parameters completely, and the description focuses on output behavior instead, which is appropriate for a parameterless read-only listing tool.

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 ('List') and a specific resource ('the collection's current curated shelves'), and then enumerates exactly what is returned: titles, descriptions, cover photos, and member details. It also distinguishes shelves from other tools by noting they are 'themed groups spanning all entry kinds,' so an agent can clearly identify this as the listing operation.

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 context: this tool lists current curated shelves and describes their composition. It does not explicitly name sibling alternatives or state when not to use it, but none of the sibling tools perform the same listing function, so the usage context is already clear enough to avoid confusion.

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 indicate mutation (readOnlyHint false) and non-destructive (destructiveHint false). Description adds that it runs in the background (~1 min) and replaces existing shelves, and suggests follow-up. Adds behavioral context beyond annotations.

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?

Moderately long but every sentence adds value; purpose is front-loaded and instructions are clear. No fluff.

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-optional-parameter tool with no output schema, the description covers the full call sequence: confirm, ask for direction, pass guidance, wait, and verify with list_conceits. An agent has everything needed.

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 already covers guidance with a description; description adds how to populate it (pass user's answer or omit for clean slate), making the parameter's use explicit.

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 (regenerate) and resource (curated shelves/conceits), and explains that it replaces current shelves. Distinguishes from sibling list_conceits which is a read-only view.

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 instructs to confirm user intent and ask for creative direction before calling, and provides the guidance parameter usage. Also tells to call list_conceits afterwards. Gives clear when-to-use context.

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

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

Beyond the readOnlyHint annotation, the description discloses meaningful behavior: the hybrid dense + ColBERT retrieval, the compact hit shape {id,title,cuisine,dietary,cookbook}, the up-to-limit cap, and the hidden trap that many recipes are unlabelled and only reachable via concept search. This is valuable context that the annotation alone does not provide.

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 but every sentence earns its place: purpose, method, examples, return shape, iteration guidance, and a critical warning. It is front-loaded with the primary action and does not waste words on redundant filler.

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's moderate complexity, full parameter schema coverage, and read-only annotation, the description is complete. It explains the expected output shape despite the lack of an output schema, gives rich query guidance, and provides clear call/re-call behavior, leaving no critical gap for an agent deciding how to invoke 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?

Schema coverage is 100%, so the baseline is 3. The description adds real semantic value for the query parameter by insisting on a rich natural-language concept over a keyword and providing concrete examples such as 'cozy cold-weather baked desserts.' The limit parameter also gains context from the phrase 'Returns up to limit compact hits.'

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 opens with a specific action and resource: 'Search the recipe collection semantically (dense + ColBERT hybrid) to gather the members of a theme.' This clearly distinguishes it from siblings like get_recipe or lookup_cuisine, and the purpose—retrieving theme members via concept search—is 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?

It gives concrete when-to-use direction: gather theme members, pass a rich concept rather than a keyword, call once per concept, and adjust the query if results are too small or too large. It also warns against relying on cuisine labels because many recipes are unlabelled, effectively steering the agent away from sibling tools like lookup_cuisine.

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

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

Beyond the annotations, the description adds meaningful behavioral details: the email always goes to the signed-in account's verified email, there is no recipient option, and the email:send permission is required. This helps the agent understand side effects and constraints beyond the basic read/write hints.

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

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

The description is three sentences with no fluff. It front-loads the core behavior, then covers exclusions, optionality, and permission requirements efficiently. Every sentence contributes useful information.

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 two-parameter tool with annotations and full schema coverage, the description covers the key operational facts: destination, lack of recipient option, optional note, and required permission. It does not describe the response/confirmation, but that is a minor gap given the tool's simplicity.

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 both parameters. The description adds minimal extra meaning, such as 'short personal note' and 'from the collection,' but does not significantly expand on the schema definitions.

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: 'Email a recipe from the collection to YOU.' It also distinguishes itself by explicitly noting there is no recipient option and that it always goes to the signed-in account's verified email, which clearly separates it from any sharing-related alternatives.

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 context for when to use the tool: when you want to email a recipe to yourself. It also provides a when-not signal by saying 'to share with others, forward it or use a share link,' though it does not name a specific sibling tool as an alternative.

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?

The description is transparent about the side effect ('replacing the current shelves', 'replace existing work'), but this directly contradicts annotations.destructiveHint=false. Since the annotation says the tool is not destructive while the description says it replaces existing work, the agent receives conflicting signals; this is an annotation contradiction.

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 front-loaded with the core purpose and the polling handoff to get_background_job, and the confirmation/cost warning is valuable. Minor redundancy ('takes minutes rather than seconds' vs 'takes several minutes'; 'replacing current shelves' vs 'replace existing work') keeps it from being 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?

For a two-parameter tool with no output schema, it supplies the needed execution context: job id for polling, duration, confirmation requirement, cost, and side effects. Missing details are minor, such as an explicit statement that conceit_reshuffle accepts no args and no failure/return shape, but the definition is strong enough for an agent to call and monitor 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 schema already documents both parameters with 100% coverage, and the description adds real meaning to the only kind value by explaining what conceit_reshuffle does, how long it takes, and that args can be omitted when none are needed. It does not further specify the args object shape, which prevents a 5.

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

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

The description starts with a specific verb and resource: start a long-running background job and get a job id to poll with get_background_job. It also names the only job kind (conceit_reshuffle) and what it does, making the tool's purpose unambiguous relative to sibling polling/list 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 gives explicit context: use this for work that takes minutes rather than seconds, expect several minutes, and confirm with the user because runs cost money and replace existing work. It does not explicitly point to a non-background alternative such as the closely named sibling reshuffle_conceits, so it stops short of a full when/when-not matrix.

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 only declare readOnlyHint=true, so the description carries the transparency burden. It adds exact-enumeration semantics, truncation signaling, node-id conventions, and metadata traversal behavior—far more than the annotations require.

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 but each segment contributes: core operation, exactness/truncation, ID format, metadata nodes, and component-node nuance. The core definition is front-loaded and the examples compress a lot of usage into a small space.

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 output schema, the description should cover result semantics; it states completeness and truncation behavior and gives node-id conventions that help parse results. It does not specify the result shape or fields beyond that, which is a small gap for a complex graph traversal tool.

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

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

Schema description coverage is 100%, so the baseline is 3. The description goes beyond the schema by explaining the '<type>:<key>' node-id format with real examples, clarifying what node_types controls, and showing valid start strings for depth and node_types usage.

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 first sentence names a specific operation—multi-hop graph walk from a start node with filters—and immediately distinguishes it with 'EXACT-enumeration tool' and concrete examples. This clearly separates it from graph siblings like find_graph_connection, get_graph_node, and get_graph_overview.

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 when-to-use context through examples and explicitly frames the tool as exhaustive enumeration. It does not name alternatives or list when not to use it, so it falls short of a full 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 declare the tool is not read-only, is idempotent, and not destructive. The description adds context that the list is synced and partners see every change, that keys must be copied exactly (never invented), and that it returns the updated list. This goes beyond annotations with useful behavioral details.

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 but well-structured, front-loading the core purpose, then usage rules, parameter semantics, alternatives, and return behavior. Each sentence conveys a distinct requirement without redundancy, earning 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?

Given the tool's complexity (4 operations, no output schema), the description covers all necessary aspects: exact operation semantics, key provenance, user consent requirement, the partner visibility effect, the alternative tool for recipes, and the return value. Nothing critical is missing.

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

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

With zero schema descriptions, the description fully compensates: it explains that check/uncheck/remove take exact keys from get_shopping_list, while add takes plain lines as spoken by the user. This gives the agent precise guidance on how to fill each parameter correctly.

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 changes the synced shopping list with specific operations (cross off, un-cross, remove, add). It also references get_shopping_list for keys and identifies add_to_shopping_list as the alternative for recipe ingredients, distinguishing this tool from its siblings.

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 says 'Call it ONLY on the user's explicit ask' with concrete examples, and directs recipe-ingredient additions to add_to_shopping_list. This gives clear when-to-use and when-not-to-use guidance, including the prerequisite of obtaining keys from get_shopping_list.

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