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

Annotations already cover the safety profile (readOnlyHint, idempotentHint, destructiveHint false, openWorldHint false), and the description adds genuinely new context: what a result contains (summary, capabilities, pricing label, how to call it) and two trust limits (directory entries are untrusted third-party text and may be stale; Haven does not pay for or vouch for paid tools). It omits rate limits and pagination behavior, but the trust disclosure is substantive.

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 purpose, then usage, then caveats. The limits sentence is long but each clause adds real information (trust, staleness, payment stance). No filler.

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?

There is no output schema, so describing the return shape ('summary, capabilities, pricing label and how to call each one') is necessary and present, and the untrusted-data warning is exactly what an agent needs before acting on results. The main gap is parameter coverage for four of six inputs.

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 only 33% (q and capability documented; limit, source, free_only, accepts_tasks are not). The description partially compensates by framing the search as 'by task or capability', which maps to q and capability, but it never explains the source enum values, free_only, accepts_tasks or limit. Baseline 3 for a partially documented schema it does not fully rescue.

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 (Searches) and resource (a catalog of tools) plus the exact scope: Haven's own free tools, indexed MCP servers, A2A agents and public-directory APIs. This distinguishes it from near-siblings like tool_capabilities, search_agents and search_skills, which cover different resources.

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 a clear trigger condition: 'Use it when you need a tool for a task you can't do with the tools you have.' It stops short of naming the alternatives (e.g. tool_capabilities for capability ids), so the routing is implied rather than explicit, but the use context is unambiguous.

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