blockrun_search

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

Annotations already declare readOnlyHint=true and destructiveHint=false, so the agent knows it's a safe read operation. The description adds significant behavioral context: the pricing model, cost estimation formula, that 'sources' defaults to web only, and that requesting an unavailable source is refused before payment. It also refers to a skill for full shape. This adds value beyond annotations, but falls short of flawless because it doesn't detail return format or pagination, though that's minor.

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 moderately long but front-loads the most critical information: the cost warning and how to control it. It uses structured sections (e.g., 'Common shape') to improve readability. Some repetition exists (e.g., cost details repeated in the first paragraph and again for max_results), but overall it's efficient and scannable. It's not as concise as ideal but appropriate given the complexity of pricing logic.

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 a complex pricing model and parameter behavior, the description covers essential usage constraints: price calculation, default values, parameter ranges, and even refers to a skill for full request shape. It doesn't explain the return value shape (no output schema), but given that it references a skill file with worked examples, it's reasonably complete. The only gap is that it doesn't explicitly state when to choose this over other search-like siblings, but the domain-specifics (news, cost) are clear enough.

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 description coverage is 100%, so the schema already explains each parameter (body, path, agent_id). The description adds further detail on the body structure, including example fields and defaults for 'sources' and 'max_results', which is useful but not strictly necessary. The description also clarifies that 'path' is reserved for future surfaces, which is beyond schema, but the schema already has a description, so the added value is moderate. Baseline 3 is appropriate.

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 clearly states the tool performs a live web and news search with AI-summarized results and citations. It specifies the endpoint path and distinguishes it from the sibling list (e.g., blockrun_exa for AI search, blockrun_polymarket for prediction markets). However, it doesn't explicitly contrast with siblings, and the name 'blockrun_search' is generic, but the description provides specific details like the price model that helps identify the tool's purpose.

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 explains when to use the tool (for real-time web and news queries) and provides explicit guidance on parameter usage, including defaults and how to cap spend by reducing max_results. It also notes the absence of an X/Twitter source, which is a useful exclusion. However, it doesn't explicitly discuss when NOT to use this tool versus alternatives like blockrun_exa for AI search; it only mentions that asking for X is refused.

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