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

The description goes far beyond the annotations (readOnly, openWorld, idempotent). It discloses the fill-check behavior (theoretical_edge_pp_at_book vs realizable_edge_pp, thin_legs, do-not-trade condition), the Jaccard similarity threshold (≥0.30), the placeholder filter (drops will-person-X / will-manager-Y, >20% returns null), and the response structure. No contradiction with annotations; the tool is read-only and non-destructive, and the description's 'signals' are informational, not executions.

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 long but highly structured, with clear section markers (SEMANTIC ANCHOR, PARTITION FILTER, FILL CHECK) that aid scanning. The first sentence states the core purpose, and every subsequent sentence adds critical operational detail. For a complex tool with no output schema, this level of detail is warranted; there is no fluff or repetition.

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?

Given the tool's complexity (3 modes, multiple internal checks, fill-risk integration, no output schema), the description is remarkably complete. It covers response structure (opportunities[], partition_check), failure modes (null arb signal, realizable_edge_pp ≤ 0), alternate tool linkage, and even the `skipped_low_similarity` field. The 0 required parameters and rich annotations lower the bar, but the description still exceeds it by explaining all runtime behaviors.

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 the schema already describes both parameters with 100% coverage, the description enriches them substantially. For `event`, it explains that it 'walks child markets, checks date-axis / threshold-axis ordering AND computes the partition_check,' with concrete slug examples. For `topic`, it details the cross-event scanning, flattening, and comparator behavior. It also notes full Polymarket URLs are accepted, which is not explicitly in the schema. This goes beyond the baseline and adds meaningful context for correct usage.

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 and resource: 'Find arbitrage opportunities on Polymarket via monotonicity violations + partition-sum checks.' This clearly distinguishes it from siblings like polymarket_edges or polymarket_fill_risk by naming the exact method and scope. It also outlines three distinct modes (trending_scan, event, topic), each with a clear purpose.

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?

Explicit usage guidance is provided: 'Call with NO args for a trending_scan, pass event for the strongest per-event partition_check, or topic for a themed cross-event scan.' It recommends `event` for a specific market, explains what cross-event mode catches that single-event misses, and even directs users to `polymarket_fill_risk` for custom sizing. This leaves no ambiguity about when to use this tool versus alternatives.

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