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

Adds substantial behavior beyond the annotations: the server never holds keys or sends transactions, the build expires after 180 seconds via `expiresAt`, builds are single-use, and specific 422 outcomes (`simulation_reverted`, `insufficient_balance` with token/required/available) are disclosed. These are consistent with readOnlyHint=true (nothing is written; the owner signs) and idempotentHint=false (a rebuild produces a different address), so there is no annotation contradiction.

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?

Despite its length, every sentence carries distinct operational value — mode requirements, custody model, expiry, single-use constraint, simulation gate, zap-in semantics, and error codes. It is front-loaded with the core purpose before the safety caveats, so an agent gets the essentials from the opening clause.

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 19-parameter tool with no output schema, it covers the critical gaps: it names `steps` and `expiresAt` in the return, explains the two-direction field discipline, and flags the zap-in denomination change. It does not describe slippage, approvalMode, or maxPriceImpact defaults beyond what the schema states, but those are adequately covered at 95% schema coverage.

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 already 95%, so the baseline is 3, but the description adds real meaning: fields of the opposite direction are refused rather than ignored, exactly one of withdrawAmount/withdrawAll must be supplied on pull, and with `tokenIn` set the `amount` is denominated in `tokenIn` decimals. That last point is a subtlety the schema alone does not convey.

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 first sentence names a specific verb (build) and resource (adjustment steps) and splits the two modes with their required fields: push needs collateralToken/amount/borrowToken/leverage, pull needs token/tokenOut plus one of withdrawAmount/withdrawAll. It is unambiguous what the tool produces (unsigned transaction steps). It never names the sibling it supersedes or pairs with (build_open vs build_adjust, quote_adjust), leaving sibling separation implicit.

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?

Strong when-to-use guidance: build immediately before signing rather than as a plan, never replay a build, and pass `steps` to `simulate` and check `ok` before asking for a signature. It also gives a retry path via `forceSwapper` when simulation fails. It stops short of naming `quote_adjust` as the tool to use when the owner is still deciding, which is the natural alternative given the sibling list.

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