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

Annotations already declare readOnlyHint, openWorldHint, idempotentHint, and destructiveHint=false. The description adds substantial behavioral transparency beyond those flags: it walks the order book ladder, returns a verdict of clean/degraded/cannot_fill, and explicitly warns that partial basket fills convert an arb into an unhedged directional position. It even names the dominant loss mode, giving the agent a clear mental model of what the tool does and what risks it surfaces.

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 description is densely packed but highly structured, using CAPITALIZED sections (SINGLE-MARKET, BASKET) and listing return fields in comma-separated groups. Every sentence adds new information—purpose, requirements, mode-specific behavior, return values, and usage guidance. Despite its length, there is no fluff; it earns its size by covering a complex tool comprehensively.

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 carries the full burden of explaining return values. It does so explicitly for both modes, listing every major field: top_of_book, vwap_fill_price, slippage_pp, shares_filled, max_fillable_usd, verdict, theoretical_sum, realizable_sum, capture_ratio, profit_usd, per-leg fill detail, thin_legs[], max_clean_notional_usd, and forced_directional_risk. It also provides practical context about when a user would need this tool. For a complex financial risk-check tool with 4 parameters and no output schema, this is exceptionally complete.

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?

Although schema coverage is 100%, the description enriches the parameters significantly. For size_usd, it explains the dual interpretation: 'max spend on buys, target proceeds on sells' in single-market mode, and 'settlement notional S (shares per leg; each share pays $1)' in basket mode. It also clarifies the side parameter semantics for both modes, including the default auto-selection in basket mode based on partition sum. This goes well beyond the schema's brief descriptions.

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 opens with a specific verb phrase: 'Realizable-vs-theoretical edge check against live CLOB order-book depth.' It clearly differentiates the tool from siblings like polymarket_arbitrage and polymarket_edges by explaining that this is a pre-trade risk/fill-simulation tool, not a signal detector. It also names the two distinct operational modes (single-market and basket) and enumerates the outputs, making the purpose unmistakable.

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 gives explicit when-to-use guidance: 'USE THIS before acting on any polymarket_arbitrage SELL/BUY-EVERY-LEG signal or any polymarket_edges trade above ~$500.' It also explains why (theoretical overround on thin books is not capturable, partial basket fills create unhedged directional risk), providing both positive and negative usage context. This clearly distinguishes it from the sibling tools that produce the signals it checks.

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