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

With annotations declaring openWorldHint=true and destructiveHint=false, the description still adds real behavioral context beyond them: web mode returns a plan for approval while native mobile 'starts immediately', native needs no URL but requires `custom_instructions`, and web requires project.main_url unless url is supplied. Missing is what the caller should do next (check_ai_crawl_status / get_ai_crawl_results) after the job starts.

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?

Three sentences, front-loaded with the core action and prerequisite, then mode-specific behavior, then the selection criterion. Dense but every clause carries information; only the em-dash aside is slightly hard to parse.

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 an 11-parameter, non-read-only, open-world job launcher with no output schema, the description omits the lifecycle: what is returned, how the plan approval step works for web mode, and which sibling tools poll status or fetch results. It covers prerequisites well but leaves the post-call path unspecified.

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?

At 36% schema coverage across 11 parameters, the description compensates only partially — it clarifies url vs. project.main_url precedence and the conditional requirement of `custom_instructions`, and implies crawl_mode is usually inferred. It says nothing about depth, pages_limit, framework, test_type, or enable_giskard_validation, which are undocumented in both places.

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 ('Start an AI crawl job for a project') and clearly splits the tool into two modes (browser web crawl vs. Native Mobile discovery). It does not differentiate itself from the sibling `start_ai_crawl_from_test_case`, which an agent could plausibly confuse with this tool.

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 an explicit trigger for the native path: 'Use it to discover real on-device element selectors when a native test case has steps the mobile flow compiler cannot resolve.' No exclusions or named alternatives (e.g., start_ai_crawl_from_test_case, import_document_and_crawl) are mentioned, so routing between siblings is only half-covered.

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