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

Annotations only supply readOnlyHint and openWorldHint; the description carries the substantive behavior: hard exclusion rules (nothing above 100% APY, nothing above 25% on sub-$1M pools, nothing with an unpublished split, small-token pairs suppressed unless includeSmallTokens), the fact that excluded entries are counted rather than silently dropped, and the output ordering contract ('lead with what staying in the token pays, then the pairs'). It also discloses the per-pool fields returned (APY split into fees/interest vs rewards, 30-day average, deposits, quality flags). That is far beyond what the annotations state.

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?

Scope is front-loaded in the first clause and the paragraph is information-dense with no filler. It is, however, built from long semicolon-and-parenthesis chains that occasionally bury the operative filter rules mid-sentence. Efficient, but a bulleted split of rules versus output shape would read faster.

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?

There is no output schema, so the description must describe returns, and it does: APY decomposition, 30-day average, deposit figures, and the set of flags. Combined with the explicit exclusion policy (which an agent needs in order to explain gaps in results) and the ordering instruction, nothing material about a single call is left to guesswork.

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%, so asset, chain, limit, minTvlUsd, and includeSmallTokens are all already documented, and the description mostly restates them indirectly ('unless asked, pairs with small tokens' maps to includeSmallTokens; the $1M and $100k thresholds interact with minTvlUsd). It adds no format or edge-case detail for individual parameters beyond the schema. Baseline 3 is appropriate.

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 opening sentence names a specific verb-of-sorts (earning yield) plus the resource (every DeFi pool DefiLlama tracks holding the token), and it explicitly enumerates the taxonomy it returns: pools that keep you in the token versus pairs that add another token. That scope is unmistakably different from every sibling listed (token_quote, perp_quote, token_report), none of which is about DeFi yield venues. An agent can route here without opening the 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?

The description supplies three literal example queries (ETH on Hyperliquid, USDC on Base, staking SOL) that define the intended invocation pattern, plus a clear statement of what it is not ('a comparison of published rates, not advice'). It never names an alternative tool to use instead, so there is no explicit when-not routing, but no sibling occupies this niche, so the practical guidance is unambiguous.

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