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

Annotations already declare destructive, non-idempotent and open-world, but the description adds substantial context beyond them: no account required, no web search or URL verification, validation rejection happens before any backend request, backend failures/timeouts can leave the outcome unknown, a download link is not a compilation-completion signal, editing does not change the app name or regenerate the icon, and omitted sections are removed. It also discloses the app-management code embedded in the id.

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?

Long, but the tool has 24 parameters and two modes, and it is organized under 'Inputs:', 'Creation:', 'Editing:' headers with the highest-value facts first. Nearly every sentence carries distinct information; the success/failure paragraph is the only place with mild redundancy against the annotations.

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?

With no output schema, the description fully specifies the return contract (structuredContent with id, download_url, edit_url, and the meaning of each), error behavior (isError=true with explanatory text), and the asynchronous compilation caveat. An agent has everything needed to call this correctly and interpret the result.

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 schema carries most parameter detail and the baseline is 3. The description adds genuinely useful cross-field rules the schema states only per-parameter: every used slot needs a nonblank title, unused slots must omit all three fields, and the edit payload represents the entire final structure with omissions treated as removals. It also points to the schemas for content formats and limits rather than duplicating them.

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 and resource ('Creates Android apps and builds APKs') and immediately distinguishes the two modes: creation versus replacing sections of a previously created app. The explicit scope limit ('initial configurable app structure, not arbitrary Android source code') prevents a major misreading of what the tool produces.

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?

Gives explicit selection criteria: previous_id is "0" for creation, a complete id from an earlier response for editing. It also states what is NOT available ('no partial-update mode or configuration-retrieval tool'), which tells the agent the payload must be reconstructed in full rather than fetched, and routes deeper customization to the web editor.

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