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

Annotations already declare readOnlyHint=true and openWorldHint=true, so the description needn't restate safety. It adds real behavioral context beyond them: the lazy/cheap-first retrieval strategy, the cost signal on errors_only, and the anti-pattern warning against full-log dumps. It does not explain truncation behavior (what happens when the tail is cut off mid-stream), which is the remaining behavioral gap. Note the annotation set itself is internally odd (destructiveHint=true alongside readOnlyHint=true), but the description's 'fetch' framing is consistent with readOnlyHint, so this is an annotation inconsistency rather than a description 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?

Front-loaded with the core action, then the workflow pointer, then the retrieval policy. Every sentence earns its place and nothing is padded; the telegraphic 'LAZY:' framing is intentional signaling rather than noise.

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?

No output schema exists, and the description covers the essential shape (a small tail of build log events, error/warning lines when filtered) plus the diagnosis-then-fix workflow and cost discipline. Only the truncated-tail case is unaddressed; everything an agent needs to call it correctly is present.

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 the schema already documents limit (default 25, max 100) and errors_only. The description still adds decision semantics the schema lacks: when to prefer errors_only=true (first, cheap) and when limit is worth raising (only if the error-only tail is insufficient). That is added meaning, not repetition.

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+resource+scope: 'Fetch a SMALL tail of Vercel build log events for a deployment.' It also positions itself in the workflow by naming code_deploy_status as the source and code_write_files as the follow-up, so an agent can distinguish it from those siblings without opening a 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?

Explicit when-to-use and how-to-sequence: diagnose a failure here, then fix with code_write_files; 'prefer errors_only=true first'; 'only pull more with limit if needed'; and an explicit exclusion, 'Never dump a full build log.' This is a complete usage policy, not implied guidance.

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