bash

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

With no annotations, the description carries the whole burden and does substantial work: it blocks, it kills the command and everything it started after `timeout` seconds (capped at 300), stdin is empty, and its environment is isolated from the user's. What it leaves undisclosed is how non-zero exit codes and stderr are surfaced, and what happens to the output if it is large — meaningful gaps for a mutation-capable shell tool, but the core runtime contract is clearly stated.

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 the verb and the environment boundary, then the scoping rules, then the alternative tool — a sensible order with little waste. The xclip/xdotool and 'stdin is empty' details all earn their place; the only slight cost is the fragmented line structure, which is stylistic rather than redundant.

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?

An output schema exists, so return formatting need not be spelled out, and the description covers the behaviors that schema cannot: blocking, kill semantics/timeout cap, empty stdin, and environment isolation. The omissions (cwd default, error/exit-code behavior) are real but secondary for a tool this well-framed.

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 0% — the three parameters have only bare titles — and the description adds real meaning only for `timeout` (max 300, process-tree kill). `command` is self-evident, but `cwd` is never explained (default value or relative-path base), which is exactly the parameter that most needs prose given the schema provides nothing.

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 action and resource (run a shell command) and defines the exact execution environment ('inside the agent desktop's environment', 'sees the agent desktop's display and D-Bus, not the user's'), which cleanly separates it from GUI siblings. The explicit pointer that `launch` exists for GUI apps further disambiguates it from the rest of the sibling list.

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?

It names concrete in-scope uses (xclip/xdotool, reading app logs, files under $HOME) and an explicit out-of-scope case with a named alternative: 'Use `launch` for GUI apps and anything long-running.' The exclusion is tied to a stated condition (blocking wait plus kill semantics), so an agent can decide without inference.

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