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

Annotations already cover readOnly/openWorld/idempotent, but the description layers substantial behavioral disclosure on top: spreads are gross not net because Kalshi's fee schedule isn't modeled; null temporal_alignment means 'could not be computed', not 'aligned'; unclassified legs are never paired; compatibility codes can be non-empty even when matched_pairs>0. No contradiction with the annotations — readOnlyHint and idempotentHint are consistent with a pure read/compute 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?

Every sentence carries real information with almost no fluff: fee disclosure, null semantics, skipped counters, per-leg flags are all load-bearing. However, it is a single dense wall-of-text paragraph — extremely long — that relies on CAPS for emphasis (TWO MODES, SAFETY FIELDS, NOT, NEVER) and lacks any bullet formatting or section breaks. Front-loaded with purpose but poorly scannable for an agent.

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?

High-complexity tool with no output schema, so the description must carry return-value semantics — and it does exhaustively: top_spreads_pp direction (Kalshi − Polymarket), compatibility_warning vs compatibility_codes semantics, per-entry flags, temporal_alignment meaning, fees_note scope, and skipped_cross_type/subtype counters. Nothing an agent would need to interpret an unfamiliar result is left unexplained.

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 100% — each parameter is already described inline. The description adds meaningful semantics beyond the schema: the topic enumeration is given a domain context (10 pre-mapped macro shortcuts that auto-fetch matching events), the explicit vs topic override relationship is clarified, and it states that both modes run the identical matcher. This is genuine value-add over a schema that merely lists allowed values.

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?

Opens with a specific verb-resource statement — 'Cross-venue spread between Kalshi and Polymarket for the same resolving question' — and goes on to distinguish the tool from single-venue siblings (polymarket_arbitrage, polymarket_edges) by framing it as a cross-venue comparison with a token-overlap matcher. The two operational modes are named explicitly with concrete examples.

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?

Gives clear operating context: what topic shortcuts do, when to fall back to explicit kalshi_event_ticker + polymarket_event_slug, and that both modes share the same matcher. It also warns that 'most pre-mapped topics return compatibility_warning today; pre-mapped ≠ tradeable.' Missing is any explicit routing against sibling tools (e.g., 'for single-venue edge analysis use polymarket_edges') — context is strong, but no named alternatives or exclusions.

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