获取阶段引导（核心工具）

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

With no annotations, the description carries the full burden and does disclose meaningful traits: the tool is stateless ('工具无状态') and performs no workflow gating ('不做流程门禁'), plus it supports rollback correction. It does not cover error behavior or confirm read-only semantics beyond 'get', but the state/gating disclosure is genuinely useful context.

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?

Purpose and the omitted-stage hint are front-loaded before the long enumeration of package contents. The 11-item list is dense but each item is a real return field with no output schema to carry them, so the length is largely earned; only minor tightening is possible.

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 two-param, zero-required, no-output-schema tool, the description compensates well by enumerating the returned package contents (what the agent would otherwise lack without an output schema) and explaining the default and stateless behavior. Only error/edge-case handling is left unstated.

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% and the schema already documents both mode (default primary/wpf) and stage (default returns overview). The description reinforces the stage-omission default behavior but adds no syntax or format detail beyond what the schema states, so the baseline 3 applies.

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 (获取) plus resource (阶段引导) and then enumerates exactly what the 'constraint package' contains (goals, artifacts, exit conditions, hard rules, etc.), so an agent knows precisely what it gets back. It also distinguishes the omitted-stage case (9-stage overview) from the per-stage case, which no sibling provides.

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?

Gives clear sequencing advice: '开场先取总览与用户建立全景' tells the agent to fetch the overview first, and '任何阶段可随时获取，支持回退修正' tells it stage fetches are unrestricted and support rollback. It does not explicitly name list_modes as the alternative for enumerating modes, though the schema does, so this stops short of a full when/when-not statement.

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