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

Annotations already declare readOnlyHint=false, openWorldHint=true, idempotentHint=false and destructiveHint=false, so the safety profile is covered. The description adds practical side-effect context: it handles two Blackboard handler types, creates the destination directory if missing, and returns resolved local paths with sizes. It stops short of describing overwrite behavior or per-file failure handling, so it is strong but not exhaustive.

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 core action and keeps each sentence purposeful: handler coverage, destination_dir guidance, return summary, and sibling routing. It is slightly longer than strictly necessary because the destination_dir sentence largely repeats the schema description, but no sentence 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?

Given the output schema already documents return structure and annotations cover the mutation/open-world profile, the description supplies the remaining decision context: handler types handled, destination behavior, and the inline alternative. It is nearly complete, with only minor gaps like overwrite semantics and error 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 coverage is 100%, so all four parameters are already documented in the input schema, including the destination_dir default and the response_format enum. The description restates destination_dir's purpose and example but adds no syntax or constraint detail beyond what the schema provides, so baseline 3 applies.

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 (download), resource (every file attached to a Blackboard content item) and destination (local disk). It also distinguishes the tool from ntulearn_read_file_content by naming the inline-inspection alternative, so an agent can tell them apart without opening both 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?

Explicitly names the alternative sibling (ntulearn_read_file_content) and the condition that selects it ('inspect the content inline rather than saving to disk'). It also gives a concrete scenario for using destination_dir, leaving no ambiguity about when to use this tool versus its sibling.

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

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

Annotations already cover the safety profile (readOnly, idempotent, openWorld, non-destructive), so the description's job is additive. It contributes ordering guarantees ('newest first'), the fan-out default, per-item courseId attribution for cross-course views, and pagination support — useful behavioral context beyond the structured fields.

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?

Multiple short sentences, front-loaded with purpose and default behavior before parameter guidance. Every sentence carries information, though the closing 'Supports pagination' is thin relative to the schema's own offset/nextOffset documentation.

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?

An output schema exists, so return-value explanation is not required. Given five all-optional parameters, full schema coverage, and annotations that carry the safety profile, the description supplies everything an agent needs: default scope, scoping mechanism, time filtering, attribution, and pagination.

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 schema already documents limit, since, offset, course_ids, and response_format thoroughly. The description adds marginal value with a concrete course_ids example and a use-case framing for since, which nudges it above 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?

States a specific verb ('Get'), an unambiguous resource ('announcements'), the scope ('across your enrolled courses'), and the ordering ('newest first'). The resource is clearly distinct from sibling tools like ntulearn_get_assignments, ntulearn_get_upcoming, and ntulearn_get_gradebook, so an agent can route without opening a schema.

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 concrete usage conditions: default fans out across all courses, pass course_ids to scope, and use since to narrow to recent items. It clearly describes when to apply each parameter but does not name an alternative tool or state exclusions, so it stops 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 cover the safety profile (readOnly, idempotent, non-destructive, openWorld), and the description goes further by disclosing merging behavior (exact-match), conflict surfacing (conflicting deadlines), provenance (source links) and a trust caveat for unverified excerpts. That is meaningful behavioral context beyond the structured 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?

Three dense sentences, front-loaded with what is aggregated before the caveats, and no filler. The internal jargon ('exact-match merging', 'documentScan') is compact but assumes familiarity with the tool's return object.

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 usefully names coverage indicators (errors/truncated/documentScan) and provenance, but omits anything about batching across course_ids, result limits, or truncation behavior tied to max_items/max_documents. Adequate but with clear gaps for a 4-parameter aggregation 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 0% and none of the four parameters (max_items, course_ids, max_documents, include_documents) are explained in prose. The mention of 'assignment documents' and documentScan implicitly gestures at the document-related flags, but limits, defaults and the meaning of course_ids are left entirely to the schema's bare types.

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 concrete resource (assignments) and specifies the sources it reconciles: gradebook, calendar and assignment pages, plus unverified deadline excerpts. It is clear what the tool returns, but it never names a sibling (e.g. ntulearn_get_upcoming or ntulearn_get_gradebook) to differentiate itself from the tools that overlap most.

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 stated condition for choosing this tool over ntulearn_get_upcoming or ntulearn_get_gradebook, even though the sibling set shows heavy overlap. The only guidance given ('never treat excerpts as confirmed deadlines') is about interpreting results, not about when to invoke 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 already declare readOnly/openWorld/idempotent/non-destructive, so the safety profile is covered. The description adds genuinely useful behavior beyond that: the recursive tree-walk model, the hasChildren drill-down signal, and pagination support. It stops short of stating rate limits or page-size guidance.

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?

Three tight sentences, front-loaded with the core action, then the two-mode rule, then pagination. No filler or restated boilerplate.

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 traversal tool with an output schema (which carries return-shape detail) and full annotation coverage, the description supplies everything an agent needs: how to start at the root, how to descend, and that results are paginated.

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 meaning: it explains the omit-vs-pass semantics of parent_id and the hasChildren precondition for valid drill-downs, which the schema alone only weakly conveys. It also reinforces the pagination intent of limit/offset.

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: 'Walk a course's content tree'. It immediately frames the tool as hierarchical traversal, which distinguishes it from siblings like ntulearn_search_course_content (search) and ntulearn_get_assignments (flat listing).

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 explicit conditions for the two modes: omit parent_id for top-level items, pass parent_id where hasChildren=true to drill in. However it never names an alternative tool (e.g. search_course_content) for when the agent already knows what it's looking for, so no exclusions are 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?

Annotations already cover safety (readOnly, idempotent, non-destructive, openWorld), so the bar is lower. The description still adds real behavioral context: the default fan-out scope, the per-column courseId attribution, and the fact that scores may be absent in some columns. It omits auth requirements and rate limits, but the added context is meaningful.

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 purpose and scope before secondary details (attribution, pagination). Every sentence carries information, though 'Supports pagination' is mildly redundant with the visible offset/limit parameters.

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, 4-parameter tool with full schema coverage, rich annotations, and an output schema, the description covers the essential call-time concerns: default scope, how to narrow, and attribution. It does not explain related-tool routing, but no sibling overlaps this resource, so the omission 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 description coverage is 100%, so the baseline is 3 and the schema already explains limit, offset, course_ids, and response_format. The description goes slightly beyond by giving a concrete course_ids example ('_123_1') and confirming pagination behavior, adding modest value over the schema text.

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 ('Get'), resource ('gradebook columns'), and scope ('across your enrolled courses'), and the 'including your scores where available' clause tells the agent what data it yields. No sibling tool competes for this resource, so the distinction is unambiguous without needing to name another tool.

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 describes the default behavior (fan out across every available course) and the condition for narrowing it (pass course_ids=['_123_1']), which is actionable guidance. It stops short of naming when-not-to-use or an alternative tool, so it lacks the exclusions that would make it 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 cover the safety profile (readOnlyHint, idempotentHint, destructiveHint=false, openWorldHint), so the bar is lower. The description still adds real behavioral context, notably that the server fans out per-course in parallel and the default window/scope, which the annotations do not convey.

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?

Three sentences, each earning its place: identity first, then the type='GradebookColumn' mapping, then defaults and the three override levers. Nothing is padded and the most decision-relevant facts are 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?

With an output schema present and pagination documented in the schema (offset/nextOffset), the description need not explain return values. It covers scope, window defaults, filtering, and fan-out behavior; only the routing distinction from sibling due-date tools is left unstated.

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% and each parameter already documents its role, defaults, and format, so the baseline is 3. The description restates the same guidance (course_ids scoping, since/until ISO-8601 override, type filter) plus one illustrative enum value, adding modest reinforcement rather than new meaning.

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 ('Get upcoming calendar items and assignment due dates across your enrolled courses') and immediately clarifies the relationship to a sibling by noting that due dates surface as type='GradebookColumn' rather than as a separate resource. An agent can separate this from ntulearn_get_assignments and ntulearn_get_gradebook without opening a schema.

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 clear operating context: default window is the next 2 weeks across all courses, and it names the three levers for changing that (course_ids, since/until, type). It stops short of stating when to prefer this tool over ntulearn_get_assignments or ntulearn_get_gradebook, which are the obvious alternatives for due-date questions.

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, idempotentHint, openWorldHint, and destructiveHint=false, so the safety profile is covered. The description adds a genuinely useful behavioral detail beyond annotations: the default result set silently excludes unavailable courses, which an agent must know to avoid misreporting a user's enrollments.

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?

Three short sentences, front-loaded with what is listed and scoped, followed by the default behavior and the override, then pagination. No filler and every clause carries 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?

With annotations covering safety, a full output schema, and 100% schema description coverage, the definition supplies the one thing structured data cannot: the default filtering semantics and the override flag. Return shape and pagination mechanics are adequately covered elsewhere, leaving only minor gaps.

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 all four parameters are already fully documented in the schema, including limit bounds, offset/nextOffset usage, the response_format enum, and the include_disabled meaning. The description's mention of include_disabled and limit/offset restates what the schema already carries, so the baseline 3 applies.

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 ('List') plus resource ('courses the current user is enrolled in') and names the platform, giving the exact scope. This is clearly distinguishable from sibling tools that act on course contents, assignments, announcements, or gradebooks rather than the enrollment list itself.

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 default-filter behavior ('returns only active/available courses') and the opt-in to include disabled ones imply when to use the flag, but no alternative tool is named and no explicit when/when-not guidance is given. Usage context is inferable 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?

Annotations already cover the safety profile (readOnlyHint, idempotentHint, non-destructive), but the description adds substantial behavior the annotations can't convey: the 1 MB MCP payload cap, ~3K vision tokens per page, per-file 25 MB / batch 40 MB / 50-rendered-page caps, and the fact that oversized or unsupported files land under `skipped` rather than failing. That is exactly the operational context an agent needs before invoking.

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 purpose and the sibling distinction before any parameter detail, then it moves to mode guidance, fallbacks, and caps. It is dense and nearly every sentence carries a distinct constraint, but at roughly 200 words it is long for a read tool and includes some redundancy with the schema's own mode/pages descriptions.

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?

An output schema exists, so return-value shape need not be explained, and annotations already carry the safety semantics. The description fills the remaining gaps a caller would hit — supported formats (Office variants with speaker notes/sheets, text-like files with HTML tag stripping), caps, the skipped-file behavior, and where to go when unsupported. Nothing material 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?

Schema coverage is 100%, so the baseline is 3, but the description adds decision rationale beyond the schema: it explains *why* text is the default ('cheap and almost always what you want for written content') and when vision is justified, and reinforces the mode+pages pairing. It is largely additive rather than duplicative, though a few points restate 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?

Opens with a specific verb+resource ('Read the content of files attached to a Blackboard content item') and immediately scopes the behavior ('returned inline (no local-filesystem hop)'). It also names the sibling it is not — 'ntulearn_download_file is for users who actually want the bytes on disk' — so an agent can distinguish the two without opening either schema.

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 states when to use this ('ask questions about lecture material') vs the alternative ('users who actually want the bytes on disk'), then gives mode selection guidance (text default, vision for diagram/equation/screenshot-heavy), page-range pairing to stay under the 1 MB cap, and a fallback path for multi-page diagram decks. Even the unsupported-binary case is routed ('fall back to ntulearn_download_file for those').

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 readOnly, idempotent, openWorld and non-destructive, so the safety profile is covered. The description adds the matching semantics (case-insensitive substring on title or description) and notes breadcrumb paths are returned, but that matching detail is essentially duplicated in the query parameter's schema description, leaving little net new behavior 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?

Three short sentences with zero filler: purpose first, matching rule second, return shape third. Nothing to trim and the most important information 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?

With an output schema present the description needn't explain return values, and annotations carry the safety profile; the description still supplies the matching rule and the recursive scope. It is nearly complete, only missing a pointer to the sibling it replaces for plain content listing.

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 (query, course_id, max_depth, max_results, response_format) is already documented in the schema with defaults and caps. The description adds no parameter-level detail beyond what the schema provides, which is the baseline expectation.

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 (search) and resource (course content tree) with the recursive scope made explicit, plus what it matches on. An agent can distinguish it from a plain listing tool, but the description never names the nearest sibling (ntulearn_get_course_contents) that an agent would otherwise confuse it with.

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

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

Usage is implied: reach for this when you need to locate an item by name/text anywhere in a course rather than enumerate the tree. There is no explicit when-not guidance and no named alternative, so the agent must infer the boundary against ntulearn_get_course_contents on its own.

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 cover the safety profile (idempotent, non-destructive, open-world), but the description adds real mechanism: manifest + checksum-based incremental logic, preservation of local edits, and the guarantee that files are never deleted. It still omits auth requirements, rate limits, and how the openWorld remote fetch behaves on failure.

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 tight sentences, front-loaded with the action and mode, then defaults, then mechanism, then the validation caveat. No filler and every sentence carries 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?

No output schema exists, but the description compensates by naming the errors and truncated result fields an agent must check. Combined with the annotations, an agent has enough to invoke correctly, though the under-documented non-required parameters leave a small gap.

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

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

Schema coverage is only 17%, so the description must carry the load. It clarifies dry_run's default and hints at truncation (max_items) and the folder layout that informs semester/destination_dir, but course_ids and max_items semantics are never explained and semester/destination_dir formats remain unstated.

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 (sync) and resource (course files) with an explicit destination scope (semester/course/original folders), plus the two modes (preview vs incremental sync). This clearly separates it from every read-only sibling like ntulearn_download_file or ntulearn_get_course_contents.

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 clear operating context: default dry_run=true, use it to preview, and inspect errors/truncated before declaring success. It does not name alternatives or exclusions (e.g., when to prefer ntulearn_download_file), so it stops short of a full when/when-not statement.

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