launch_game

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

Annotations already declare readOnlyHint=false, openWorldHint=true, idempotentHint=false, destructiveHint=false, and the description adds real context beyond them: a refusal path for already-running processes, automatic session-context switching with an index rebuild, the context_switched return field, and a context_warning instead of guessing when the exe's game can't be inferred. It even explains the failure mode it prevents (index/runtime image mismatch with ok:true). It does not cover what happens to the previously active session or timing/return payload shape, keeping it below 5.

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 core action and the key refusal rule, but the session-context-switching behavior is explained twice in near-identical paragraphs, and the parenthetical about 140 vs 143 images is over-detailed. The Args block is efficient, but the body could be roughly halved without losing information.

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 launching/mutating tool with no output schema, the description covers prerequisites (session identity follows exe), failure modes (already-running process, uninferable game), and names the return fields the agent should inspect. The only real hole is the undocumented `preferred_module` parameter and what a successful launch returns beyond context_switched.

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%, so the description carries the burden and largely succeeds: `exe` documents the empty-string default (use the session's identified main program) and `args` documents Windows-style splitting, quote handling for spaced paths, and that backslashes are not escapes. The third parameter, `preferred_module`, is never mentioned anywhere, leaving one of three parameters completely undocumented.

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+resource (launch game + inject probe before initialization) and immediately differentiates from the sibling inject_process by explaining why launching first is stronger. The scope is unambiguous and an agent can pick it over inject_process without opening a schema.

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?

Explicitly names when to use it (intercepting one-shot initialization logic before objects exist), the alternative it beats (manually starting the game then inject_process), and a hard refusal condition (same-name process already running, which wastes 60s). That is when/when-not/alternative all in one place.

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