Agent Cost Log

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

With no annotations, the description carries the full behavioral burden, and it does disclose meaningful traits: scope depends on API key presence, and history depth is tier-gated (free 7 days, Pro 30). It omits that this is a read-only operation, the return shape, and any pagination behavior, so gaps remain.

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 terse sentences, purpose front-loaded, with zero filler. The aphoristic style ('With a key: everything on that key') is compact but slightly telegraphic, leaving some phrasing to be parsed.

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

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

For a two-parameter read report with no output schema and no annotations, the description covers the essentials: what is broken down, the auth-dependent scope, and the tier-based window. Missing details are modest — default value for days and confirmation of read-only, non-mutating 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 only 50% — run_id has no schema description — and the description compensates by explaining that run_id selects a single run and only applies without an API key. The tier limit restated for days ('Free shows 7 days, Pro 30') adds meaning beyond the schema's '1-30 (free max 7)'.

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

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

The first sentence names a concrete report and its breakdown dimensions (per run, per tool, per model, per day), which is far more specific than a tautology. However, it never distinguishes this report from the sibling tools log_costs, usage_and_upgrade, or weekly_report, leaving overlap unresolved.

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

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

It gives conditional invocation rules — pass run_id when unauthenticated, otherwise the whole key is covered — which is useful context. But there is no when-to-use guidance relative to the sibling cost/report tools, so an agent must infer which report to pick.

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 discloses the 500-event limit, free (watch-only, would_alert) vs Pro (single daily webhook alert) behavior, and that it "never blocks or changes your agent." It stops short of stating idempotency, retention, or auth requirements, which is a real gap for a write-accepting endpoint.

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?

Dense but front-loaded: purpose first, then input modes, then caps/alerts, then the non-blocking guarantee. Nearly every clause adds information, though the single wall-of-text paragraph makes predicates like the OTel field mapping easy to skim past.

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, nested-object tool with no annotations and no output schema, the description covers input modes, caps, alerting tiers, and top-level return values. It leaves return shape (beyond would_alert and totals) and error/permission behavior unstated, but is otherwise sufficient to call correctly.

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

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

Schema coverage is only 63%, and the description compensates meaningfully: it maps OTel attributes to outputs (gen_ai.usage.input_tokens/output_tokens/gen_ai.request.model read; trace id becomes run_id), clarifies cost_usd vs tokens+prices ("nothing is guessed"), and labels run_cap_usd/daily_cap_usd as optional caps. The nested otel object and the missing alert_webhook semantics keep it below a 5.

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

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

Specific verb+resource ("Log what your agent's model and tool calls cost") plus the side effect of returning running totals for the run and today. It is clearly distinguishable in function from a pure reporting tool, but it never names cost_report or the other siblings, so sibling differentiation is left to inference.

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 two input modes (events vs otel) and details cap/alert behavior, which implies when to reach for this tool. However, it gives no explicit when-to-use-this-vs-cost_report guidance or prerequisites, so the agent must infer the boundary from names alone.

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, so the description carries the burden. It adds useful context: free tier of 500 calls/day, not metered. However, it doesn't state whether the call itself counts toward quota, rate limits, or whether the upgrade link is static/dynamic.

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 short, front-loaded sentences with no waste. The primary output is stated first, followed by the supporting quota context.

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-param, no-output-schema read tool, the description covers what is returned (usage and upgrade link) and the quota context. It lacks detail on return format, but that's minor given the simplicity.

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

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

Zero parameters, so baseline is 4. The description correctly indicates the scope is implicit (your key or IP), needing no input.

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 ('Show') and resource ('today's usage for your key or IP') plus a secondary output (the upgrade link). It's clearly distinguishable from siblings like budget_status and weekly_report which deal with budgets/reports.

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?

Implied usage via 'today's usage' and the free-tier context, but no explicit when-to-use or when-not-to-use versus siblings like budget_status or weekly_report. The agent must infer that this is a quick self-check rather than a budget query.

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 a decent job: it declares the operation read-only, states that it 'sends nothing' (no side effects or data egress), and flags the auth requirement ('needs a free API key'). Missing are finer traits such as caching/freshness of the 7-day window or any rate limits.

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 compact sentences: the first front-loads what is returned, the second covers the safety/auth profile. Every clause 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?

With no output schema, the description compensates by enumerating the returned sections, and it closes the annotation gap with read-only/auth statements. Minor gaps remain: the boundary definition of the 7-day window (rolling vs. calendar, timezone) is 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?

The tool takes zero parameters, so there is nothing to disambiguate and the baseline is 4. The description correctly spends no words on input syntax.

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?

Concrete verb+resource: a plain-English summary of the last 7 days of spend, with the exact contents enumerated (total, top runs, tools, models, busiest day). The fixed 7-day scope implicitly separates it from siblings like cost_report, but no sibling is named, so the differentiation is left to inference.

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 name and the 'last 7 days' framing, but the description never states when to prefer this over cost_report or usage_and_upgrade, nor any when-not condition. It is adequate routing-wise only because the scope is narrow and obvious.

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