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

With no annotations, the description carries the full behavioral burden. It adds useful context about flakiness and the explicit baseline-refresh intent, but says nothing about what happens when the comparison fails (exception vs. returned diff), whether it blocks, or what permissions/environment it requires.

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 correctly front-loaded in sentence one and nothing is buried. The middle paragraph drifts into rhetorical justification ('how visual testing earns its reputation for flakiness') rather than information the agent needs to call the 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 5-parameter assertion tool with no annotations and no output schema, the definition covers purpose and region rationale but omits the failure contract — the single most important thing an agent needs when invoking an assertion. Adequate but with a clear 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 100%, so device, region, baseline, tolerance and update_baseline are already documented in the schema. The description reinforces why region matters and clarifies the intent of update_baseline, but adds no format, range or interaction detail beyond the schema, 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?

The opening sentence gives a specific verb (compare) and resource (current screen or region) against a baseline image, so the operation is unambiguous. It does not, however, distinguish itself from the sibling run_and_assert, which an agent could plausibly reach for on the same task.

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 explains why region cropping is preferable to whole-screen comparison, which is real usage guidance for the region parameter, and it frames update_baseline as the deliberate refresh path. It never says when to call this tool versus run_and_assert or the record/explore siblings, so routing is left to inference.

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 does substantial work: it enumerates the artifacts collected (before/after screens, Maestro capture, focused window, logcat slice, elements at failure), discloses that an attribution heuristic runs, that a JSON plus HTML bundle is written, and the notable behavior of returning diagnosed: false on pass. It does not mention auth/permission needs or rate limits, so not a 5.

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

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

Front-loaded with the core action, then the evidence list and the pass-case return behavior. The final sentence about diagnosed: false is arguably long-winded but it conveys a genuinely non-obvious behavior, so it 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?

There is no output schema, so the description must explain returns — and it does highlight the diagnosed: false case, though it doesn't describe the success-path bundle structure or the shape of the attribution result. For a 5-parameter, side-effecting diagnostic tool with no annotations, that remaining gap keeps it at 4.

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 (env, flow, device, package, timeout) is already documented in the schema. The description adds no format or syntax detail beyond that, so the baseline of 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?

The description states a specific verb and resource: run a flow and, on failure, gather evidence to attribute fault. It clearly distinguishes itself from a simple assertion runner by emphasizing diagnosis and artifact production. It doesn't explicitly name the sibling tools it competes with (e.g. run_and_assert, run_parallel), so it stops short of a 5.

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 want post-failure diagnosis — but there is no explicit 'use X instead when Y' guidance against run_and_assert, run_parallel, or explore_and_record. An agent can infer the context but is left to reason about alternatives 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?

No annotations are provided, so the description bears the full burden. It does disclose a genuinely important behavioral trait — the deliberate double execution of steps when both capture_assertions and validate are on — but omits file-writing side effects (output_dir), device acquisition behavior, and what happens on validation 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?

Front-loaded with the core purpose, followed by a short rationale paragraph. Only minor flourish ("a hypothesis, not a test") that still conveys the reason for the double run, so the text is nearly all 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?

For an 8-parameter, no-annotation tool, the description covers purpose and the notable double-run behavior but says nothing about return values, error handling, or where the generated file lands beyond the output_dir schema entry. Adequate, with clear gaps for an agent deciding how to call and interpret 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 description coverage is 100%, so a 3 is the baseline. The description earns a point above baseline by explaining the interaction between two parameters — capture_assertions plus validate causes two runs — which the schema documents individually but never jointly.

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: generating a replayable Maestro flow from a described exploration, then verifying it runs. That is far more informative than the bare name, though it never names how it differs from siblings like run_and_assert or run_parallel.

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 context is implied by the exploration-to-replayable-flow framing, and it explains the condition under which steps run twice (capture_assertions and validate both enabled). However, it gives no explicit when-to-use-this-instead guidance relative to run_and_assert or run_parallel, which are the obvious alternatives.

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 burden. It describes what is reported and distinguishes two failure modes, which is useful, but it does not state that the operation is read-only, what permissions are needed, or any side effects. Adequate but with gaps.

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, with the core purpose front-loaded and each sentence adding value: the first states what it does, the second gives the usage trigger, and the third clarifies the diagnostic value. 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 zero-parameter diagnostic tool with no output schema and no annotations, the description covers purpose, usage trigger, and the kind of result to expect. Nothing an agent needs to call it correctly 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 tool takes zero parameters and the schema description coverage is 100%, so there is nothing to document. The baseline for a 0-parameter tool is 4.

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: 'Verify the toolchain' and enumerates the report contents ('which tools work, which are impaired, what to install'). Clear but does not explicitly differentiate from diagnostic siblings such as debug_failure, so it falls short of a 5.

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 trigger: 'Run this first when anything behaves unexpectedly.' It does not name alternatives or say when not to use it, so it lacks the explicit alternative routing that would earn 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?

With no annotations, the description carries the full burden and does well: it defines the `usable` state, and explains that `unauthorized` and `offline` devices still appear but cannot run flows. It does not cover permissions, pagination, or refresh cadence, so it is not fully 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, followed by a short usage instruction and a compact state glossary. Every sentence earns its place with no 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?

There is no output schema, so the description must convey returns; it does so by naming health and lease state and defining the state values. Return structure (fields per device) remains unspecified, which is a 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?

The tool takes zero parameters, so the baseline is 4 and there is nothing further for the description to document.

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 ('List every connected Android device and emulator') plus the fields returned (health and lease state). An agent can immediately distinguish it from siblings like health_check or run_parallel.

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 prescribes timing ('Call this before planning any parallel work') and explains why it is the sole source of truth for usable device count. It does not name an alternative or a when-not condition, so it falls short of the top bar.

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 burden. It does disclose genuinely non-obvious behavior: the flow result is always reported and assertions are deliberately skipped after a failed flow. However, it says nothing about device allocation/side effects of running a flow on a pooled device, nor what the result payload looks like.

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, followed by two short sentences that justify the skipping behavior. Efficient overall, though the final causal clause is slightly more verbose than strictly needed.

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 flow-executor/asserter with no output schema and no annotations, the description covers failure semantics but omits return shape and device-pool side effects. The rich input schema compensates on the input side, leaving a moderate output/behavior 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 description coverage is 100%, so every parameter including the nested ElementAssertion fields is already documented. The description adds no parameter-level meaning (e.g., how 'flow' paths resolve or how 'device' interacts with the pool), 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 and resource: 'Run a Maestro flow, then check assertions against the screen it left behind.' This clearly distinguishes it from a plain runner, and the assertion-on-final-screen framing hints at separation from assert_visual. It stops short of explicitly naming which sibling to prefer, so it lands at 4 rather than 5.

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 the conditional behavior rule (assertions skipped when the flow failed) but never states when to choose this tool over run_parallel, assert_visual, or debug_failure. Usage must be inferred from the action itself.

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 does so well: round-robin scheduling, ~N/M round count, devices leased for the whole duration, guaranteed release on exit including failure (so a crash cannot strand a device), and per-device failure isolation. These are exactly the operational traits an agent needs before invoking a resource-leasing 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?

Front-loaded with the core action and scope, then two short paragraphs covering scheduling and return/failure behavior. Every sentence carries information an agent would otherwise have to guess; nothing is restated.

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

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

There is no output schema, but the description compensates by describing the return payload (per-flow results, wall-clock time, sequential estimate, speedup) and the failure-reporting model. Combined with the 100%-covered input schema, an agent has everything needed to call and interpret this 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 schema already documents env, flows, devices, and timeout, including the busy-device-skipped rule. The description's mention of distribution and leasing adds context about how devices are used but not new per-parameter semantics. Baseline 3 is appropriate.

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 gives a specific verb ('Run'), resource ('Maestro flows'), and the distinguishing scope ('in parallel across leased devices'). That scope cleanly separates it from the sibling run_and_assert, which implies single-flow execution. No ambiguity about what the tool does.

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

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

The parallel/multi-flow scenario is implied by 'Run several Maestro flows in parallel', and the device-pool behavior is explained, but there is no explicit when-to-use vs when-not guidance. It never states when to prefer run_and_assert or a single-flow path, nor does it note any prerequisites for leasing devices. Implied usage only.

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