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

With annotations already declaring readOnly=false/openWorld=false/idempotent=false, the description adds substantial context beyond them: execution happens on the owner's device, the owner must click 'Allow' in the studio, they can watch and stop it, input is synthetic (not trusted events), pointer lock is simulated, and `request_id` makes retries idempotent. It also enumerates the `stan` status codes (brak_urzadzenia, karta_w_tle, zatrzymane, etc.) and the evidence/dowod_urzadzenia outcome.

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 opening sentence front-loads the purpose and the target/session model follows logically. It is dense and long, with a status-code list and evidence paragraph that tax readability, but for a 9-parameter tool with a nested step timeline most sentences carry real 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?

No output schema exists, so the description must carry the return-value burden — and it does: screenshots as image content, game state, `stan` codes, `dowod_urzadzenia` evidence, and how klyo_game_diagnostics surfaces it. Combined with the permission/Allow prerequisite and session lifecycle, an agent has everything needed to invoke it correctly.

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 100%, so the baseline is 3, but the description meaningfully complements it by explaining the step grammar: steps run on one timeline of at most 3000 ms, each starts when the previous ends or at `at_ms` for parallelism, coordinates are fractions 0-1, up to 40 steps and 4 screenshots, and it lists the fields per step type. This composition semantics goes beyond the per-field schema text.

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+resource ('Really play your game and see it: you send input steps, you get screenshots (image content) and the game state') and explicitly frames the mode as remote-on-owner's-device. It also distinguishes itself from siblings by tying its target to 'the same version klyo_game_files reads' and routing failures to klyo_game_diagnostics.

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 session lifecycle is spelled out clearly ('The first call opens the game, next calls continue it, `restart` reloads it, `close` ends the session'), and the error path names an alternative ('Errors from this session: klyo_game_diagnostics with `sesja` = `sesja.znacznik`'). It does not state explicit when-NOT-to-use conditions versus siblings like klyo_game_files, so it stops short of a 5.

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