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

With annotations only covering the safety profile, the description carries real weight: it discloses the permission required on the library's project, that viewers are read-only, the plan gating for SharePoint/OneDrive providers, the host restrictions on those URLs, the feature toggle for task linking, and the dedupe behavior ('same address already on the task returns the existing link with created: false'). The dedupe note sits in mild tension with idempotentHint=false, but it is disclosed rather than hidden, so the agent is not misled about mutation effects.

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 core concept and the two-way branch are front-loaded in the first sentence, and every subsequent clause carries a rule (permissions, provider subsets, host constraints, dedupe). It is a single dense paragraph of ~200 words whose readability would improve if the library and task branches were separated into bullets, but there is 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?

For a 6-parameter mutation tool with no output schema, the description is nearly complete: it covers both modes, their preconditions, provider legality, and the 'created: false' return signal. It does not say what a fresh link returns (document/link identifier) or what happens when a matching link already exists in a library rather than a task.

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 83%, so the baseline is 3, but the description goes further than the schema by mapping which provider values are legal in which mode (SHAREPOINT/MICROSOFT_DRIVE only for taskId, and https-only there vs http(s) for libraries) — something the single flat enum cannot express. The externalFileId parameter is only described in the schema, not the description, which is the remaining gap.

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?

Opens with a specific verb and resource ('Attach an external URL as a zero-byte link') and immediately splits the two targets (libraryId vs taskId), so an agent knows exactly what is created. The 'zero-byte link' phrasing implicitly distinguishes it from file-upload siblings, but no sibling (e.g. attach_file_from_url, upload_task_attachment) is named to make the choice explicit.

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 hard selection rule ('exactly one of the two'), then describes the distinct outcomes of each mode and the precondition for each (project.content.create with viewers read-only; Microsoft-files plan; task links enabled; task attachment permission set). It stops short of naming a competing tool, so the agent still has to infer when to prefer this over attach_file_from_url.

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