Doco Rebuild Summary

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

With only thin hints in annotations (readOnlyHint=false, idempotentHint=true), the description carries the behavioral burden and does so richly: it discloses the sync-versus-async contract per generator, the mandatory follow-up polling via doco_summary, the obsolescence condition when the source changes or the summary is pinned, and the guarantee that stale results will not overwrite. This is consistent with all annotations — idempotentHint=true aligns with the non-overwrite and idempotency_key behavior — and no contradiction exists.

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 compact — two sentences, front-loaded with the core action '重建摘要' followed by generator-specific contracts. Every clause carries distinct information (execution mode, return behavior, staleness policy, overwrite guarantee) with no filler or restatement of the tool name. The semicolon-chained structure is dense but readable, losing a point only for the slightly run-on feel.

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 rich input schema (100% coverage, enums, idempotency key), the idempotentHint annotation, and an existing output schema, the description only needs to fill the behavioral contract — which it does thoroughly via sync/async mode, staleness conditions, and non-overwrite guarantees. The residual gaps are selection criteria between deterministic and model generators and the exact meaning of '可追溯 fallback', but neither blocks a correct invocation.

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 the schema already documents all six parameters and the baseline is 3. The description adds only indirect value: the obsolescence condition implies meaning for base_source_version, and the pinned-summary behavior relates to target scope, but no parameter-level syntax or format is elaborated. This is acceptable because the schema handles the burden.

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 the verb+resource pair '重建摘要' (rebuild summary), clearly identifying a write-rebuild operation on a summary. It further distinguishes itself from the sibling query tool doco_summary by stating that a model generator requires a follow-up query via doco_summary(job_id). However, it never names doco_save_summary, its closest writing sibling, so an agent must infer the distinction between rebuilding and saving without explicit guidance.

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 gives concrete, actionable workflow guidance: for the model generator it explicitly states that only an async task is created and the agent must subsequently call doco_summary(job_id) to retrieve the result, which routes an agent correctly between two sibling tools. It provides clear context for the deterministic vs model split but lacks an explicit statement of when to prefer this over doco_save_summary or criteria for choosing between the two generators.

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