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

Annotations already establish readOnlyHint=true, but the description adds real behavioral context: the shape of the returned step records, the CACHED/UNCACHED cache-state vocabulary, and the fact that error text and log availability are surfaced per step. It does not mention auth requirements or pagination limits, but page_size/page_token are covered in the schema.

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?

Front-loaded with the action and the returned field list, followed by the routing hint and the RPC identifier. Dense but every clause carries information; the trailing "Depot: build.v1 BuildService/GetBuildSteps" is somewhat ornamental but cheap.

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 usefully enumerates the fields an agent will receive and the enum-like cache states, which is the main completeness gap it had to fill. It omits ordering/grouping semantics for steps and any note on pagination behavior, but is otherwise sufficient for calling the tool correctly.

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 description coverage is 100%, so all four parameters (project_id, build_id, page_size, page_token) are already documented in the schema. The description only references digests as an output to hand off, adding no new parameter-level meaning. Baseline 3 applies when the schema does the heavy lifting.

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?

States a specific verb and resource ("List the BuildKit steps of a container build") and immediately enumerates the returned fields (name, digest, timing, cache state, error, log presence). This clearly separates it from siblings like depot_get_build (the build itself) and depot_get_build_step_logs (the log content).

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 an explicit workflow: "Find the failing step here, then pass its digest to depot_get_build_step_logs." This names the alternative tool and the exact condition (and the parameter value) that routes the agent to it, leaving nothing to inference.

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