Mined links graph

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

Annotations only declare readOnlyHint=true, so the description carries most of the burden and does so well: it discloses that results are persisted batch output only, that edges are urn-keyed and outlive catalog rebuilds (so counterpart fields can be null), that results are grouped by kind and by direction, that `runs` supplies provenance, and that output is capped per kind with an honest note. The only gap is that it never states the shape of the empty/error response or whether pagination beyond the cap exists.

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 key framing is front-loaded in the first clause, but the body is one sprawling sentence threaded with nested parentheticals and heavy jargon ('a star per refrain whose detail carries the gram and score its count', 'meet: ref · root [shelf]'), which forces re-reading. The content is nearly all load-bearing, but the packaging could be split into discrete sentences or bulleted kind descriptions without losing anything.

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 the description must describe returns itself, and it does: grouping by kind, both direction values with their meanings, counterpart fields resolved to title/language/license_class/source with null semantics, provenance via `runs`, and bounding behavior. For a read-only tool with a fully documented 3-parameter schema, nothing an agent needs in order to call it correctly or interpret the result 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%, so the baseline is 3; the description earns an extra point by clarifying that the cap applies per kind and that an 'honest note' accompanies truncation, which the schema's 'Maximum edges returned per kind' does not convey. The restricted-material gate is mostly explained in the schema rather than the description, and 'PRESERVE the license fields when quoting' is output guidance rather than parameter guidance.

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 opening clause names a specific verb and resource ('Batch-mined cross-reference edges touching a passage/document urn') and immediately enumerates the four edge kinds with their producers, so an agent knows exactly what data comes back. It also names the sibling it is not ('for on-the-fly discovery use nabu_parallels'), which separates it cleanly from nabu_parallels and nabu_cognates.

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?

It explicitly routes the agent elsewhere when this tool is wrong ('for on-the-fly discovery use nabu_parallels') and pre-empts the most likely misinterpretation: 'an empty result means no batch has covered this urn, NOT that no parallel exists.' That is an explicit when-to-use, when-not-to-use, and how-to-read-the-result set of instructions.

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