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

Adds substantial context beyond the read-only/idempotent annotations: one row per player per game at the median closing line, results split by season, a 95% interval, and the explicit disclaimer that it describes the past and recommends nothing. It also discloses the 25-credit cost, which the annotations do not cover. A 4 rather than 5 because it doesn't say what happens on thin samples or whether low-count buckets are dropped.

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 purpose, then output shape, then grain, then the sibling pointer and cost, so an agent can stop reading early. The example sentence and the 'recommends nothing' disclaimer each earn their place, though the description runs long for a tool whose schema is fully documented.

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 17-parameter aggregation tool with an output schema, the description covers purpose, aggregation grain, output contents (hit rate, interval, implied and fair probabilities, per-season split), historical scope, and cost. Nothing an agent needs to call it correctly or interpret the result 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?

With 100% schema coverage the schema already documents all 17 parameters, so the baseline is 3. The description still adds meaning by tying filters to the worked example ('receiving yards overs for tight ends on home underdogs' maps to market, side, position, team_role) and by clarifying that the line filter is the median closing line and that fair-probability filters refer to the chosen side. It does not explain season/week interaction or the side default.

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: aggregate a backtest of a prop hypothesis, comparing how often the side hit versus the market-implied probability. It immediately distinguishes itself from the sibling get_prop_history by grain (aggregate rate vs one player game by game), so an agent can route without opening either 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?

Explicitly frames the use case ('test an idea ... without pulling every row'), names the alternative for the drift case ('For one player game by game use get_prop_history'), and bounds history to finished games since 2023. Only a hard when-not list is absent, which is minor.

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