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

Annotations already declare the read-only, idempotent, open-world, non-destructive profile, so the bar is lower. The description adds genuinely new context: the local API key authentication requirement and the fact that it reads from whichever environment (sandbox or production) is configured, which affects which credentials/balance the agent sees.

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 tight sentences with the core action front-loaded, followed by the prerequisite and return shape. 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?

For a zero-parameter read tool with no output schema, the description covers the essentials: auth prerequisite and the return content (currency and balance). It could note the environment-selection implication more explicitly, but nothing critical is missing.

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

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

The tool takes zero parameters, so the baseline is 4. The empty schema leaves nothing for the description to clarify, and it correctly adds no parameter discussion.

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 ("Read account balance") and scopes it to the configured sandbox or production application. It is clearly distinct from the write-oriented siblings (at_send_sms, at_send_airtime) and the config reader (at_get_config), though it never explicitly names or contrasts them.

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 by the read-only balance purpose, and the prerequisite "Requires a local API key" is stated, but there is no explicit when-to-use vs. when-to-use-something-else guidance relative to the sibling tools.

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 establish readOnly, idempotent, non-destructive and closed-world behavior, so the bar is lower; the description still adds useful context by promising no network access, no credential requirement, and no sensitive identifiers in the result. It does not contradict openWorldHint=false — quite the opposite, it reinforces a purely local read.

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. Every clause (local, safety limits, no network, no credentials, no phone numbers) carries distinct information.

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

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

For a zero-param, read-only, annotation-backed tool with no output schema, the description supplies enough to call it correctly and to expect what comes back (no sensitive identifiers). A brief note on what the safety limits actually look like would make it fully complete, but nothing essential 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 is a closed empty object, so there is no parameter semantics to explain and the baseline of 4 applies. The description correctly does not invent inputs.

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 ('Read') and resource ('local configuration and safety limits'), which is clearly distinct from the sibling tools that fetch balances or send SMS/airtime. An agent can identify what this returns without opening anything else.

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 'without network access, credentials or phone numbers' phrasing implies this is a safe, offline inspection path, which hints at when an agent would reach for it, but no explicit alternatives or conditions ('use this instead of X when Y') are named.

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 the destructive/non-idempotent/open-world profile, and the description adds genuinely new operational context: server-side policy gates, a required user approval step, a configured currency/cap constraint, and an explicit 'never auto-retry' contract. These are exactly the traits an agent needs before triggering an irreversible send.

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 with the default/opt-in behavior front-loaded, followed by prerequisites and the retry contract. No filler; each clause carries distinct operational information.

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

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

For a destructive, non-idempotent mutation with no output schema, the description covers mode selection, gating, approval, amount format, currency/cap, and retry behavior. It does not describe the failure/response shape, a minor omission given annotations already flag the safety profile.

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 67%, with dryRun and currencyCode already documented in-schema. The description reinforces the dryRun semantics and adds a constraint not present in the schema — the configured cap on amounts — plus the 'exact recipients' targeting expectation, adding real value over the structured fields.

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 specific verb+resource (validate/preview vs. purchase/send airtime) and makes the default vs. dryRun=false distinction explicit, so an agent immediately knows this tool can both simulate and execute. It does not explicitly differentiate itself from the sibling at_send_sms, which is the only gap.

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 clearly states the default behavior (preview/validate) and the condition that flips it to a real send (dryRun=false plus server gates), and adds that user approval is required. This is strong context, though it never names a sibling or an explicit when-not-to-use-this alternative.

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

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

Annotations already declare destructiveHint=true, readOnlyHint=false and non-idempotency, but the description adds genuinely new context: messages are billable, server-side policy gates still apply, user approval is required, retries are forbidden, and dry-run output is not a price quote. That is exactly the operational detail an agent needs before firing a destructive action.

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, each earning its place, with the default behavior front-loaded ahead of the escalation condition and the dry-run caveat last. 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?

Covers billing, approval, gating and retry policy for a destructive write tool with no output schema; the schema handles recipient/message constraints. It could say more about partial-batch failure behavior, but it is complete enough to invoke correctly.

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

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

Schema description coverage is only 50% (dryRun and senderId are documented; recipients and message are not), so the schema does not carry the full load. The description clarifies dryRun's intent and the approvals it implies, but adds nothing about recipient format, batch size, or message constraints.

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 set (validate/preview/send) against a concrete resource (SMS) and immediately disambiguates the default mode from the sending mode. An agent can tell this is the messaging tool and not the airtime sibling without opening the 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 the exact condition that escalates to a real send (dryRun=false with server gates enabled) and an explicit human-approval prerequisite plus a no-auto-retry rule. It stops short of naming alternative tools such as at_send_airtime, so it earns a 4 rather than a 5.

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