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

With no annotations, the description carries the full disclosure burden. It does reveal a non-obvious behavioral trait — that supplying pairing_code/pair_port triggers a pairing step before connecting, with the values sourced from a specific device UI screen. However, it says nothing about authentication requirements, timeouts, retry behavior, or whether pairing persists across connections.

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, no filler, and the primary action is front-loaded before the conditional pairing detail. 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?

No annotations and no output schema, so the description must cover safety and result behavior on its own. It handles the complex pairing path well but omits failure modes, whether pairing is one-time, and what a successful or failed connection returns, which leaves an agent under-informed for a network-mutating call.

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%, so the description must compensate, and it does explain the two least obvious parameters (pairing_code, pair_port) and their origin. It adds nothing about host format (IP vs hostname), port, or what happens when pairing fields are omitted alongside a non-5555 port, leaving roughly half the parameters thinly specified.

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 ('Connect to a device over Wi-Fi'), which is unambiguous against the sibling set of device-control tools like tap, swipe, and screenshot. It does not explicitly contrast itself with any alternative connection method (e.g., USB), but none exists in the sibling list, so the gap is minor.

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 a clear conditional: for Android 11+ wireless debugging, pass pairing_code and pair_port to pair first. That is actionable when-to-use guidance tied to a specific platform and flow. No when-not guidance or mention of prerequisites such as same-network reachability.

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

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

No annotations are provided, so the description carries the full disclosure burden and it says nothing beyond the one-line purpose. It does not state that this is a read-only query, what happens when no device is connected, or whether it errors versus returning empty — all relevant for an ADB-backed tool.

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

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

A single, front-loaded sentence with zero filler — exactly the right size for a one-parameter query tool. It is terse but not vague, so brevity does not cost it here.

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 read tool with a fully documented parameter, the description is minimally sufficient. With no output schema, though, it should at least characterize the return value (e.g., package name string, null when unknown) and the no-device case, which it omits.

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 the single optional 'serial' parameter is fully documented in the schema, including its default-resolution behavior. The description adds no parameter detail, which is acceptable at this coverage level but earns no bonus.

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') and resource ('the foreground app package'), so an agent immediately knows the tool returns the currently-active package. It does not, however, reference related siblings like list_apps or ui_dump to disambiguate in a crowded toolset.

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

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

There is no guidance on when to use this versus list_apps, ui_dump, or open_app, nor any stated prerequisites (e.g., that a connected device is required). The agent must infer usage entirely from the name and schema.

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

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

No annotations are provided, so the description carries the full behavioral burden. It does not disclose whether the value is physical vs. usable resolution, how rotation/orientation affects the result, or the shape of the return value, so an agent cannot anticipate the output beyond 'some number in pixels'.

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

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

A single short sentence with no filler, front-loaded with the verb and resource. Nothing is wasted, though there is also nothing beyond the minimum.

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 read-only getter with 100% parameter coverage and no output schema, the definition is minimally viable. It still leaves open what the return value looks like (e.g., width/height pair) and which resolution (physical, usable, orientation-adjusted) is reported.

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 single 'serial' parameter is fully documented in the schema (100% coverage), including its default behavior when omitted. The description adds nothing about the parameter, but per the baseline rule full schema coverage justifies a 3.

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 screen resolution') plus the unit ('in pixels'), so the operation is unambiguous. It does not reference any sibling, but none of the siblings (tap, swipe, screenshot, etc.) overlap with this retrieval, so the risk of misselection is low.

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 statement of when to call this versus alternatives such as screenshot or ui_dump, and no prerequisites or context (e.g., before computing tap coordinates). Usage is only implied by the tool name.

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

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

With no annotations, the description must carry the full behavioral burden, and it does disclose the accepted key vocabulary and the numeric-keycode fallback, which is meaningful. It says nothing about error behavior for unknown keys, whether the event is blocking, or device-state requirements, so the disclosure is partial.

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?

Purpose is front-loaded in the first four words, and the long value list that follows is functional rather than filler. The semicolon-heavy enumeration reads slightly like a dump, but every token earns its place as valid input.

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 device command with no output schema, the description plus schema covers what an agent needs to construct a call. The remaining gap is behavioral (invalid-key handling, device prerequisites), which matters somewhat given the absence of annotations.

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 only 50%, and the undocumented parameter is `key` — the description compensates by enumerating 25 valid names plus the numeric keycode option, information absent from the schema (no enum). It does not add anything about `serial`, but the schema already documents that one.

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 ("Send a key event") and enumerates the accepted key identifiers, which makes the tool's scope unambiguous against siblings like tap, swipe, and type_text. It stops short of explicitly naming or contrasting an alternative, so it is clear rather than sibling-differentiating.

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 statement of when to prefer this tool over tap, type_text, or swipe, and no prerequisites (device connected, active app) are mentioned. The key-name list hints at remote-control navigation, but that is value guidance rather than usage guidance.

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

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

With no annotations, the description carries the full burden. It does disclose one real behavioral fact — that third-party packages are returned by default, so system apps are excluded unless requested — which is useful and not derivable from the schema. However, it says nothing about the requirement for an ADB-connected device, output format, ordering, or pagination.

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

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

A single front-loaded sentence with no filler; the scoping detail is placed immediately after the action. It is efficiently sized, though brevity here edges toward under-specification rather than true conciseness.

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 three-parameter, zero-annotation, no-output-schema tool, the description is too thin. It omits any explanation of the 'filter' parameter, the ADB device dependency implied by 'serial', and what the listing actually returns.

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 33%: 'serial' is documented in the schema, while 'filter' and 'include_system' have no descriptions. The parenthetical '(third-party by default)' does explain include_system's default behavior, but 'filter' remains completely unexplained in both schema and description, so the description only partially compensates for the coverage gap.

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 clear verb+resource ('List installed packages') and clarifies the default scope with '(third-party by default)'. It reads distinctly from siblings like list_devices or current_app, though 'packages' versus the tool name 'apps' is a minor terminology wobble.

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 when-to-use guidance, no exclusions, and no mention of alternatives such as open_app (to launch) or current_app (to inspect the foreground app). The agent must infer the context entirely from the name.

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

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

With no annotations, the description carries the full burden. It does disclose one non-obvious behavioral trait — results are filtered by the ADB_SERIAL allowlist when that variable is set — but says nothing about return shape, error behavior, or whether device state (offline/unauthorized) is included.

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

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

One tight sentence with the core action front-loaded and the conditional filtering caveat parenthetically appended. No padding.

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

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

For a zero-param, no-annotation, no-output-schema discovery tool, the description covers what is listed and the one environment-dependent modifier. The only gap is the absence of any hint about returned fields, which is minor for this tool type.

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 takes zero parameters, so there is nothing for the description to clarify; the baseline for a 0-param schema 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 and resource: listing connected ADB devices. No sibling tool (list_apps, current_app, ui_dump, etc.) shares this resource, so an agent can route to it without ambiguity.

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 only implied: you call this to discover which devices are targetable before running tap/swipe/open_app. There is no explicit when-to-use statement, no mention of alternatives, and no note about behavior when no devices are connected.

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

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

No annotations are provided, so the description carries the full behavioral burden. It discloses nothing about the default 800ms hold, whether the call blocks until the gesture completes, or what side effects (context menus, drag start, selection) the press triggers.

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 single clause is front-loaded and free of filler, so nothing is wasted. But at four words it is under-specified rather than truly concise, leaving essential context for the caller to infer.

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 interaction tool with no annotations and no output schema, the description omits the gesture's effect, duration semantics, and device targeting. An agent has to fall back entirely on the schema and prior knowledge of ADB-style long presses.

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 only 25%: x and y have no descriptions at all. The phrase 'at coordinates' weakly implies x/y are screen coordinates, but the description adds no units, coordinate origin, or range information to compensate for the gap.

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 (long-press) and target (coordinates), so the basic action is identifiable. However, it does little to distinguish itself from the sibling tap or tap_element beyond the word 'long', and gives no indication of what a long-press accomplishes on this device.

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

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

No guidance on when to use long_press instead of tap, swipe, or tap_element. There is no mention of prerequisites (e.g., needing a connected device) or of the conditions that make a long press the right gesture.

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

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

With no annotations, the description carries the full behavioral burden, yet it says nothing about what 'launch' entails (foregrounding an installed app, waiting for startup, behavior on failure, error returned if the package is missing). Only the surface action is 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?

A single front-loaded sentence with no filler, which is appropriate in shape. It is arguably too terse for a tool with unannotated side effects, but there is no wasted text.

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

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

No output schema, no annotations, and an undocumented 'package' parameter leave significant gaps: return value, failure modes, device-selection defaults, and whether the call blocks are all unstated. The one-line description is insufficient for this complexity.

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 50%: 'serial' is documented in the schema while 'package' is not. The description partially compensates by clarifying that 'package' means the Android package identifier rather than a display name, but it adds no format examples or constraints, so it is only baseline-plus.

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 ('Launch an app') plus the identifier type ('by package name'), which clearly separates it from siblings like list_apps and current_app. It does not explicitly name those alternatives, but the action is 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?

There is no guidance on when to use this versus current_app or list_apps, and no stated preconditions such as the app needing to be installed or a device being connected. The agent must infer all usage 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?

No annotations are provided, so the description carries the full behavioral burden. It discloses only the PNG format; it does not say whether the image is returned inline, written to a path, or how large/complex screens are handled, nor whether any permissions or a connected device are required.

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 words plus a parenthetical — front-loaded and free of padding. It is arguably under-specified rather than bloated, but as pure conciseness it is efficient.

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, likely read-only capture tool with one optional well-documented parameter, the description is minimally adequate. The PNG hint is useful, but with no output schema the description should clarify the return shape (inline image vs. file path), which it omits.

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 single optional 'serial' parameter is fully documented in the schema (100% coverage, including the default-resolution behavior), so the baseline of 3 applies. The description adds nothing about parameter meaning beyond the schema.

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

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

States a specific verb+resource combination ('Take a screenshot') and adds the output format (PNG). No sibling tool covers screenshots, so no explicit differentiation is needed, but the description never mentions the device-targeting dimension the sibling set (serial-based ADB tools) implies.

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

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

There is no guidance on when to use this tool, no prerequisites (e.g., an active device connection), and no mention of alternatives such as ui_dump for structured UI inspection. The agent must infer usage entirely.

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

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

No annotations are provided, so the description carries the full behavioral burden. It discloses that the gesture takes ms milliseconds, but says nothing about whether the call blocks until completion, how duration influences fling/scroll momentum, what happens with out-of-bounds coordinates, or what is returned.

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

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

A single tight sentence that front-loads the gesture and parameter mapping with zero filler. It is arguably under-specified rather than verbose, which is not a conciseness failure.

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 6-parameter device-input tool with no annotations and no output schema, the description leaves the coordinate system, timing behavior, error cases, and return behavior entirely to guesswork. Only the most literal parameter mapping is covered.

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 only 17% (just serial), but the description compensates by naming x1,y1 as the start point, x2,y2 as the end point, and ms as the duration. It still omits coordinate units, the valid range for coordinates, and that ms defaults to 300 with a 1-10000 bound.

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 (swipe) and resource (a gesture between two coordinates) with the duration parameter folded in. It is distinguishable from siblings like tap and long_press, though it never names them, and the coordinate space (pixels vs normalized) is left unstated.

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

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

There is no guidance on when to prefer swipe over tap, long_press, or a sequence of taps, nor any prerequisite or context about the target screen state. The agent must infer usage entirely from the tool name.

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

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

No annotations provided, so the description carries full burden. It doesn't disclose whether the tap is synchronous, what happens if coordinates are out of bounds, whether serial selection falls back, or what response is returned. Only the coordinate-based nature is conveyed.

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?

Extremely concise and front-loaded: a single clause that states the action and its operand. No wasted words.

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?

Adequate minimum for a simple 3-param gesture tool with no annotations and no output schema. However, for a device-interaction tool it should mention coordinate space, device targeting, and synchronous 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 33%; only 'serial' is documented in the schema. The description adds no meaning beyond the schema for x/y, but the schema itself already constrains x/y as integers with minimum 0 and documents serial. Baseline 3 for partial coverage where the schema does most of the work.

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?

Clear verb+resource: 'Tap' at 'pixel coordinates'. Distinguishes from sibling tap_element (which likely taps by element selector rather than coordinates), though it doesn't explicitly say so.

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

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

No guidance on when to use tap vs tap_element, long_press, or swipe. The sibling tap_element is the obvious alternative and isn't mentioned. Usage is only implied by the name.

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

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

No annotations are supplied, so the description carries the full behavioral burden. It usefully discloses matching semantics (substring, case-insensitive) and where the tap lands (element center), but is silent on failure behavior when nothing matches, whether it waits/retries for the element, and whether it scrolls to find off-screen elements.

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

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

A single front-loaded sentence with no filler; the lookup criteria precede the action, which is the right ordering for an agent deciding whether it can use this tool.

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

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

For a side-effecting UI interaction tool with no annotations and no output schema, the description covers the happy path only. Missing failure/ambiguity semantics (no match, multiple matches beyond the schema's `index` hint, off-screen elements) leave an agent guessing about edge cases.

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 50%: `index` (match selector) and `serial` (device routing) are documented in the schema, while `text` and `resource_id` are bare. The description compensates for `text` by specifying substring/case-insensitive matching, but leaves `resource_id` matching semantics (exact vs substring) undefined and never explains that `text` and `resource_id` are alternative selectors.

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 pair (find + tap), the resource (element), and the two lookup keys (text/content-desc or resource-id), plus the tap target (center). It implicitly separates itself from the coordinate-based `tap` sibling but never names it.

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 only implied: an agent can infer this tool is chosen when it knows the element's text or resource-id rather than its coordinates, but the description never says when to prefer `tap`, `long_press`, or `ui_dump` instead, nor any prerequisite like the app already being on screen.

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

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

With no annotations, the description carries the full burden and adds one genuine behavioral constraint: '(printable ASCII)'. That usefully caps accepted input beyond the schema. However, it omits failure behavior, whether the field is cleared first, and permission/device requirements.

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

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

A single tight sentence with the actionable resource front-loaded and no wasted words. It is efficient, though slightly under-specified rather than optimally informative.

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 input-mutation tool with no annotations and no output schema, the definition is minimally adequate. It covers what the tool does and an input constraint, but leaves prerequisites (field focus) and error handling unaddressed given the absence of structured safety signals.

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 50% (serial is documented, text is not). The description partly compensates by characterizing the text parameter as printable ASCII, adding meaning beyond the schema, but says nothing about the 1000-char maxLength or the serial parameter's role.

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?

Clear specific verb+resource: 'Type text into the focused field'. The 'focused field' scope distinguishes it from tap/tap_element, but it never names those siblings or key_event explicitly, so sibling differentiation is only implied.

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

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

The phrase 'focused field' softly implies a prerequisite (something must already be focused) but the description never states when to use this over key_event or tap_element, nor what to do if no field is focused. Usage is inferable but not explicit.

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

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

With no annotations, the description carries the full burden. It usefully describes the return shape (clickable/labelled elements with bounds and center coordinates), which is genuine behavioral information. However, it omits whether this is a read-only operation, any performance/pagination characteristics, or device-selection 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?

A single front-loaded sentence that leads with the verb and progressively narrows to the output format. No filler or redundancy.

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

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

For a no-annotation, no-output-schema tool, the description covers the essential return content (element names, bounds, center coordinates) that an agent needs to use the result. It leaves gaps on device-selection and read-only semantics, but is largely sufficient for a simple dump operation.

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?

One optional parameter (serial) with 100% schema description coverage, so the schema already documents the default-device fallback behavior. The description adds no parameter meaning beyond the schema. Baseline 3 is appropriate when the schema does the heavy lifting.

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 (Dump) and resource (UI hierarchy), then further specifies the output content: clickable/labelled elements with bounds and center coordinates. This clearly distinguishes it from screenshot (visual capture) and the interaction tools like tap_element, though no sibling is named explicitly.

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

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

No when-to-use or when-not-to-use guidance is provided. The description never explains how this relates to alternatives like screenshot or tap_element, nor the workflow context (e.g., inspect-then-act). Usage is only inferable from the purpose statement.

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