grok_steer

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

No annotations are provided, so the definition carries the full behavioral burden. It does disclose a valuable trait – that appended instructions take effect only in '后续处理' (subsequent processing) and are not applied retroactively – but says nothing about return behavior, whether the call blocks, or what happens if the task has already completed.

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?

A single sentence with no wasted words; the action and target come first, and the timing constraint follows immediately.

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 no annotations, and the schema coverage is 33%, so the definition needs to carry more. It omits the return value, the blocking behavior, and the precondition that the task ID must refer to a live task, leaving real gaps for a mutation-style tool on a sibling set with six steering and lifecycle primitives.

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 only 33%: source_id is the only documented parameter, and its fallback behavior by host (Codex metadata vs. Claude session ID) is spelled out only in the schema. The definition implies the roles of the two required parameters – id identifies '执行中的任务' and prompt is the '指令' – so it adds some meaning beyond their bare string types, but it does not compensate for the coverage gap.

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 definition pairs a specific action ('追加指令' – append an instruction) with a specific target ('执行中的任务' – an executing task), plus a timing qualifier ('在 Grok 后续处理时生效'). This clearly separates it from grok_spawn (create) and grok_interrupt (stop), though it never names those siblings explicitly.

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?

By saying the task must already be '执行中' (executing), it implies when to call it rather than when not to. It gives no explicit routing advice – for example, that you should use grok_spawn when the task is not started or grok_collect when it has finished – so the agent must infer which sibling to use.

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