estimate_agent_cost

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

With no annotations supplied, the description carries the full burden and does substantial work: it explains the underlying model (every step re-sends system prompt, tool definitions, prior outputs, so input tokens grow), lists what the response contains (per-run and per-month cost, per-step breakdown, context-growth share, caching savings, cross-model comparison), and discloses defaults. It does not state that the tool is purely computational/side-effect-free or whether pricing data is cached/stale, which would complete the behavioral picture for a cost-estimation 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?

The purpose and the differentiating insight are front-loaded in the first two sentences, followed by outputs and then usage examples, which is a sensible order. The prose is dense but each sentence carries information; the long 'Returns ...' enumeration is the only stretch that could be trimmed.

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 11-parameter tool with a nested hosting object, no annotations and no output schema, the description is appropriately thorough: it covers the modeling assumption, the return payload and the default scenario. Remaining gaps are minor — no statement of where pricing data comes from or how fresh it is, and no guidance on how the cross-model comparison set is chosen beyond 'popular current models'.

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 documents each parameter, including the nested hosting fields. The description adds meaningful framing by explaining why token parameters accumulate per step and that the defaults model a typical 8-step tool-using agent, but it offers no parameter-specific guidance beyond the schema (e.g. how compare_models interacts with provider, or typical magnitudes).

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 ('estimate what one run of an AI agent costs, and a month of runs') and immediately distinguishes the modeling approach from naive per-call math via the conversation-resend explanation. This differentiates it from siblings like compare_token_pricing or compare_llm_vs_selfhost, which price tokens rather than whole agent runs.

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 'Use for:' list gives four concrete triggering questions ('what will my agent cost at 50k runs a month', 'is Sonnet or Haiku cheaper', caching savings, AgentCore/LangGraph budgeting), which is strong positive routing. It also clarifies the default agent shape and tells the agent to pass measured token counts for real figures. What is missing is explicit negative guidance — no sentence says to prefer compare_token_pricing for pure per-token comparison or analyze_ai_spend for observed spend.

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