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

Annotations already declare readOnly/openWorld/idempotent, and the description adds rich behavioral context on top: the exact findings-packet shape (verbatim evidence + confidence + source + fetched_at + pipeworx:// citation), explicit gaps[] for unanswered facets with 'never invented', contradictions[] behavior on standard/thorough, the hop field and citation_uri resolvability guarantee, semantic excerpting instead of head-truncation, latency expectations (15-60s, thorough up to ~90s), and the paid-plan requirement for thorough. This substantially exceeds what annotations alone convey.

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 (~250 words) but every sentence earns its place: auth gate and fallback, tool identity, decomposition behavior, return contract, usage guidance, exclusions, depth iteration semantics, citation resolvability, excerpting policy, and latency. It is front-loaded with the most operationally critical gate (ACCOUNT REQUIRED + fallback). The only cost is that it reads as a dense wall of text with no paragraph breaks, but for a tool this behaviorally complex the density is justified.

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?

This tool has no output schema and is complex (parallel routing, multi-hop depth modes, contradictions, citations), yet the description independently explains the full return contract, gaps[] and contradictions[] semantics, hop field, citation_uri behavior, tuning knobs, latency profile, and auth tiers. Everything needed to select and invoke the tool correctly, interpret results, and choose a depth value is present. Nothing essential is missing.

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% and the schema already documents both params thoroughly — including the depth enum semantics ('quick=3', 'standard=3 (default; adds a gap-recovery hop...)', 'thorough=6 (paid...)'). The description largely restates the same question semantics ('Broad/multi-part is fine — decomposition is the point') and depth behavior, adding only marginal extra context like the signup gate. Schema does the heavy lifting, so baseline 3 is correct.

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 names a specific verb and resource ('Grounded multi-source research across Pipeworx's 1533 STRUCTURED data sources... in ONE call') and immediately distinguishes itself from siblings: 'this is NOT open-web search', and it names what it is not relative to ask_pipeworx and ask_pipeworx_grounded. An agent can unambiguously tell this apart from every sibling without opening any schema.

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 when-to-use ('Best for broad/multi-part questions over structured data') with concrete examples, and explicit when-not-to-use with named alternatives: 'For a single lookup use ask_pipeworx instead', 'Not for current-events/news topics... that's ask_pipeworx_grounded', and the auth gate 'If you are not signed in, use ask_pipeworx instead'. All routing conditions are spelled out, nothing left to inference.

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