Check training progress by job and enterprise ID, returning status, step, loss, reward curves, checkpoints, and usage snapshot for long-running task polling.
训练进度查询——status/step/loss/reward 曲线/断点/用量快照（长任务轮询入口）。
| Name | Required | Description | Default |
|---|---|---|---|
| last_n | No | 曲线窗口（可选——尾部 N 条 progress 事件，缺省全量） | |
| train_job_id | Yes | 🔴 训练任务标识 | |
| enterprise_id | Yes | 🔴 企业标识（隔离分区依赖） |
Changes observed during successful MCP inspections.
v0.1.0Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries the full behavioral disclosure burden. It characterizes the tool as a query ('查询') and a polling entry ('轮询入口'), strongly implying a read-only, repeatable operation, and it lists the snapshot-style information returned: status, step, loss/reward curves, checkpoints, and usage snapshot. It does not mention polling cadence, rate limits, or error handling, but the core behavior is clear.
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 a single dense sentence that front-loads the core purpose, then packs the relevant output categories and usage context with no filler. Every segment earns its place: the query type, the returned data, and the polling-entry role.
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 no output schema, the description sufficiently summarizes return content and identifies the polling use case. Together with a fully documented schema and required enterprise_id for isolation, an agent can correctly select and invoke this tool. It could add more detail about terminal conditions or polling behavior, but it is complete enough for a status-query tool.
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 three parameters, including the optional tail-window semantics of last_n. The description itself adds little parameter-level meaning beyond reinforcing that this is about progress and polling, so the baseline of 3 is appropriate.
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 names a specific verb and resource: '训练进度查询' (training progress query), and enumerates concrete output categories (status/step/loss/reward curves, checkpoints, usage snapshot). It also labels itself as the '长任务轮询入口', which clearly separates it from sibling train_* operations like train_submit, train_list, or train_report.
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 explicitly says this is the entry point for polling long-running training tasks, giving an agent clear context on when to use it. It does not name exclusions or contrast itself with specific alternatives such as train_report or train_list, so it stops short of a perfect 5.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.