diagnose

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

Annotations already declare readOnlyHint=true, idempotentHint=true, and destructiveHint=false, so the safety profile is covered. The description adds meaningful behavioral context: it discloses that the tool reads logs, returns a structured verdict with findings/remediation, and includes a missing_logs field that tells the agent how to obtain absent data. It also implies a time-window default (7d) and per-device/source scoping. This goes beyond the annotations without contradicting them.

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 longer than average, but every section earns its place: the one-sentence summary is front-loaded, the symptom list is a compact reference table, the Args block is a clear parameter guide, and the Returns block lists the output fields. It uses bullet-like formatting and avoids redundancy. The length is proportional to the tool's complexity, and the structure makes it scannable.

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 tool with five parameters and ten symptom variants, the description covers all necessary aspects: what it does, which symptoms it handles, how to set time windows and filters, and what it returns (verdict, findings, details, missing_logs, advice). An output schema exists, so the description needn't detail every return field, but it names them all. Nothing an agent needs to select or invoke it correctly is missing.

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 documentation for parameters (coverage 0%), so the description carries the full burden. It does this thoroughly: 'symptom' is explained with a list of ten valid ids and what each checks; 'since' and 'until' specify ISO or relative formats with a default; 'source' and 'device' clarify scope and that 'omit for all sources'. This is exactly the semantic enrichment an agent needs and goes well beyond the bare schema.

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 ('Run a symptom playbook') and a clear resource ('logs'), and states the output ('verdict with evidence'). It also enumerates ten distinct symptom playbooks, which sharply distinguishes it from sibling tools like search_logs or summarize_errors by establishing it as a high-level diagnostic orchestration rather than a low-level log query. The purpose is unambiguous and fully covers the tool's role.

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 symptom list itself is a strong implicit usage guide: it tells an agent exactly which symptom ids map to which diagnostic scenarios, and the description notes it 'read[s] the right logs' per symptom. However, it never explicitly says when NOT to use this tool or points to an alternative (e.g., 'for raw log inspection use search_logs'). The context is clear enough for most cases, but explicit exclusions are missing, so it falls just short of a 5.

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