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

Annotations are minimal (readOnlyHint=false, openWorldHint=true, destructiveHint=false), so the description carries the burden of behavioral disclosure. It rises to the occasion: async execution and polling with research_id, 'refused' status for unsafe requests before work is queued, clarification flow for missing decision details, and the market semantic ('available in <market>' does not imply stock guarantee). All credit beyond what annotations provide.

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 description is long (~140 words) but each clause earns its place. It front-loads the most important info first (purpose and sibling distinction), then lifecycle, then safety refusals, then market nuance, and closes with an example. It forms a single dense paragraph that could benefit from a line break or two, but it's appropriately scoped for the tool's complexity.

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 tool with this much complexity (async background jobs, polling, safety refusals, clarification flows, market nuance, personas), the description covers the critical call paths thoroughly. The output schema handles return-value documentation. The main gap is not addressing the other siblings (compare_products, get_product, etc.) or edge cases like timeout/error states, but what's here is cohesive and complete enough to call 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 schema already documents each parameter. The description adds genuinely useful semantics on top: the market parameter biases discovery but not stock, research_id enables polling for the async run, and the example answer key ({'budget': '500 USD', 'market': 'US', 'category': 'espresso machine'}) clarifies the answers object shape. Valuable, though some of this is also in the schema's descriptions.

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 states the specific function: 'research_shopping is the full verified-research job', and immediately distinguishes its scope from sibling find_products, calling it the 'quick candidate-list tool'. It also specifies the resource (considered purchase) and the output (verified research with per-claim receipts). An agent can unambiguously tell this from the alternatives.

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 description explicitly contrasts with find_products (quick candidate lists vs. full research), which is the single most likely point of confusion, and outlines the polling lifecycle (running → poll with research_id). However, it doesn't address the other six siblings or give an explicit 'use this when...' rule beyond the find_products dichotomy, so some routing decisions are left to inference.

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