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

Annotations only declare readOnlyHint=false and destructiveHint=false; the description carries substantial behavioral weight beyond that. It discloses the non-guessing policy ('Nothing is guessed'), the unmapped list for unmatched target fields, the 'inferred': true flag on each mapped field in the stored mapping tree, the draft-never-publish guarantee, and the 'Requires edit access' permission. There is no contradiction with the annotations — a draft write is consistent with readOnlyHint=false.

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 verb and resource are front-loaded in the first sentence, with supporting behavioral detail (non-guessing, unmapped list, draft-not-publish) following. It is a single dense paragraph of roughly 150 words, which is justified given 9 parameters, mutually exclusive modes and nested objects, but it could be tightened or structured slightly.

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?

Given the complexity — 9 parameters, nested objects (sample, targetSample), mutually exclusive input modes, an enum, and no output schema — the description covers the operational essentials well: both input paths, mapping semantics, the unmapped list, the inferred flag, the draft pipeline, and the access requirement. It does not spell out the exact return shape, but it describes the stored mapping tree and unmapped output, which fills most of what an output schema would convey.

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 100% with every parameter described, so the baseline is 3. The description adds workflow-level meaning on top: it consolidates the two mutually exclusive input modes (targetDocument vs targetSample+targetUrl), references the servers[]-wins precedence over targetUrl, and ties targetFilename's YAML/JSON role to targetDocument. This relational gloss exceeds what the individual schema descriptions provide.

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+resource ('Generate a draft endpoint under a line') and describes a concrete mechanism: infer the inbound body schema from a sample, read the target's method/URL/fields from an OpenAPI document or bare sample + targetUrl, and propose a same-named outgoingMapping. It explicitly contrasts itself with filtering/publishing siblings by declaring it 'always writes a draft (never publishes)', with downstream flow through preview_line_draft, dry_run_endpoint, config_diff and publish_line_draft, so an agent can separate it from the similar-named draft/publish tool family.

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 clear context on what inputs drive the tool (targetDocument XOR targetSample+targetUrl, deliveryMode, draftRevision) and states the exclusion 'never publishes', which routes agents to publish_line_draft for the publish step. It names the downstream pipeline tools. It does not explicitly spell out when this should be used instead of create_endpoint, but the draft-vs-live distinction is strongly implied.

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