finish_build

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

With no annotations provided, the description carries full responsibility for behavioral disclosure. It does so thoroughly: it reveals that components are re-measured with `measure` and never trusted from prior tool self-reports, that the build is only exported if every component is a valid shape AND the count matches begin_build's declaration, and that failures are reported but not saved. This gives the agent a complete picture of the tool's side effects and decision logic.

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 well-structured and front-loaded: a one-sentence summary, a usage paragraph, a detailed behavior paragraph, and an Args section. Despite its length, every sentence contributes unique information about when to use the tool, how verification works, and what conditions trigger a save. The organization makes it easy for an agent to parse quickly while retaining all necessary detail.

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 and the absence of annotations or schema descriptions, the description is remarkably complete. It covers the workflow context (after begin_build), the verification method (measure), the success/failure criteria (valid shapes and count match), the outcome (exported or reported with reason), and parameter semantics. An output schema exists, so the lack of explicit return-value detail is acceptable; nevertheless, the description still mentions that failures are 'reported back to you with the specific reason,' offering a partial preview of the response.

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?

The input schema provides zero description coverage (only field titles), so the description must compensate. It includes an Args section that explains doc_name as 'The document passed to begin_build for this build' and component_names as 'Names of every component you built (should match, in count, what you declared in begin_build's expected_components).' This adds crucial context about how the parameters relate to the build lifecycle and the count-matching requirement, which 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 description opens with a specific verb and resource: 'Close out a build: re-verify every claimed component, and save it as a reusable example if it actually checks out.' This clearly states the tool's function and outcome, and distinguishes it from siblings like begin_build or measure by framing it as the finalization step that only saves verified builds.

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 explicitly states when to use the tool: 'Call this after begin_build and after you believe the build is complete.' It also explains the verification and export conditions, implying that it should not be used before the build is complete or when components are known to be invalid. This provides clear contextual guidance relative to the build workflow.

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