fix_plan

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

With no annotations, the description carries the full burden and does well: it discloses result ordering (most likely first, default all, typically 4-5), that each cause carries steps/parts/safety/stop conditions, and — importantly — that strength labels are experience-based tiers, not statistical probabilities. It does not state that the tool is read-only or anything about latency/caching, but nothing here suggests mutation.

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 a bracketed label, then what it returns, then the ranking caveat, then when to use it, then the token knob — a sensible order with little filler. It is dense and packed into essentially one paragraph, which makes the token-saving note easy to miss but wastes nothing.

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?

There is no output schema, and the description compensates well by describing the response shape (ordered causes, per-cause steps, parts, safety, stop conditions, tier labels). The only completeness gap is the disagreement between the stated default result size and the schema's max_causes default, which could cause the agent to over- or under-request.

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 baseline is 3. The description adds the purpose behind max_causes/max_steps (token saving), which the schema lacks — but it also conflicts with the schema by claiming the default returns all causes, 'usually 4-5', while max_causes is documented as defaulting to 3, so an agent gets two different answers about the default result size.

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 gives a specific verb and resource: '取某症状的分步方案' and then enumerates the exact payload (causes ranked by likelihood, full action steps, parts, safety warnings, stop-and-call-a-technician conditions). It is clearly distinguished from a bare ranking tool by the step/parts/safety content, but it never names a sibling (e.g. cause_ranking or diagnose) to make the boundary explicit.

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?

'回答「我该怎么修、最可能是什么、先做哪一步」用它' gives concrete trigger phrasings, and the max_causes/max_steps note explains the only real usage variation (token-saving). There is no when-not guidance — nothing says when to prefer cause_ranking or stop_check instead.

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