Start persistent process

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

Annotations already declare destructiveHint=true, openWorldHint=true, and readOnlyHint=false. The description adds genuinely new behavioral context beyond them: the session-ID return model and the RACCOON_ENABLE_SHELL=1 feature gate, which signals the tool may be unavailable. It does not describe resource costs of a long-running process, but the added gate/return info is valuable.

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?

Two tight sentences with the purpose and the return contract front-loaded and the prerequisite appended. Every clause earns its place with zero filler.

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 6-parameter, destructive, open-world tool with no output schema, the description gives an adequate high-level picture (what it returns, feature gate) but leaves all six parameters undocumented and says nothing about interactions with the process lifecycle siblings. It is minimally viable but has clear gaps.

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 0% across 6 parameters, so the description must carry parameter meaning, yet it names none of them (command, cwd, runner, deviceId, deviceName, maxLogBytes). The runner enum and the maxLogBytes byte-budget cap are especially non-obvious and unexplained.

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 (Start) and resource (long-running shell process), plus its key return contract: a session ID for incremental output reads. This clearly conveys what the tool does, though it does not explicitly differentiate itself from the close sibling 'start_process' or 'shell_run'.

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?

Usage is implied by 'long-running shell process', so an agent can infer it is for persistent/interactive sessions rather than one-shot shell_run. However, there is no explicit when-to-use-vs-alternatives guidance, and it never names the sibling tools (read_process_output, force_terminate) it pairs with.

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