chemdraw_decorate_scope

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

Annotations are all default-false (readOnlyHint: false, destructiveHint: false), so the description carries the full burden, and it delivers extensively. It discloses that the operation writes to a NEW copy with source preservation, that 'uncertain native writes stop without retry' (fail-fast behavior), that output is native editable CDXML/SVG/PNG with a layout audit, and that human visual review is still required. This level of failure-mode and side-effect disclosure is exemplary.

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?

Four dense sentences carrying distinct payloads: core action, group invariant, label/non-behavior, and output/failure semantics. The core action is front-loaded and every sentence earns its place. The density is high enough that an agent must parse carefully, but nothing is redundant and the organization parallels how the tool will actually be used (inspect → decorate → verify).

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 tool's complexity (7 params, a nested groups structure, no output schema, all-false annotations), the description is unusually complete. It documents the required invariant (exclusive ownership, nonoverlapping top-to-bottom bands), label constraints (empty allowed, nonempty needs measured free space), and failure behavior. The only notable gap is the absence of any description of the layout audit's return format, which matters since there is no output schema to fall back on.

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 0%, so the description must compensate, and it does for most parameters: 'optional native rounded shadow frame' covers frame, 'dotted group dividers' covers separators, 'New absolute output directory only' covers output_dir, and 'groups {label,fragment_ids,caption_ids} must own every source fragment and caption once' gives the groups parameter real meaning. Two parameters (pixels, document_id) receive no description-level semantics, and expected_source_token is only referenced obliquely via 'Inspect current IDs/source token first', so compensation is strong but not complete.

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 opens with a specific verb+resource: 'Decorate an existing flat scope in a NEW copy' with concrete outcomes (rounded shadow frame, dotted group dividers). The phrase 'No automatic chemical classification or molecule reordering' draws a boundary that helps separate it from analysis/build siblings. It does not, however, name any sibling explicitly, and with 40 sibling tools (build_scope_job, grid_document, polish_document) some ambiguity remains about where decoration ends and other visual-editing tools begin.

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?

The description gives clear preconditions ('Inspect current IDs/source token first') and constraints ('New absolute output directory only'), which imply when the tool is appropriate. It also signals what the tool will NOT do ('No automatic chemical classification or molecule reordering'), a weak when-not boundary. However, it never names an alternative tool or states explicit conditions for choosing a sibling instead, leaving usage routing to inference.

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