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

Rich behavioral disclosure beyond annotations: racing behavior and its consequences, method-specific failure modes (Samsung IMEs intercepting . / @ / _ as autocomplete/action shortcuts), auto-fall-back to IME for non-ASCII, and clearFirst reading current length from UI to send exactly matching backspaces. All of this adds substantive context in absence of annotations, which carry zero information here.

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 critical ordering warning first, then method options, then optional params. The content is dense but deliberately organized. Slightly long, though every sentence earns its place — the race-condition example and Samsung IME caveats are high-value behavioral details, not padding.

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 5-param text-input tool with no output schema and no annotations, the description is comprehensive: covers all three methods with reliability and device-specific caveats, explains the interplay with focusing tools, documents optional params, and handles edge cases (non-ASCII fallback, Samsung IME interceptions). No obvious gaps remain for a correct agent invocation.

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?

While schema coverage is 100%, the description adds material meaning beyond the schema: it explains WHY keys is the default (decomposes ASCII to keycode events), describes the non-ASCII fall-back mechanism, clarifies what clearFirst actually does (reads length from UI and sends matching backspaces), and defines pressKey behavior for common type-then-submit flows. This is genuinely additive guidance, not schema restatement.

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?

Description uses a specific verb+object ('Type text into the focused field on Android'), clearly stating the resource (focused Android field). It distinguishes from sibling tools like device_tap/device_tap_by_text by explicitly warning to call this AFTER those complete, and the method options differentiate it from ios_send_keys, session_send_keys, etc.

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 states when to use the tool (after device_tap/device_tap_by_text completes), warns against parallel invocation with a concrete observed failure case (banking-app login race condition with concatenated username+password), and provides per-method usage guidance including which to prefer ('keys' default most reliable) and Samsung IME caveats for shell and ime methods.

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