novel_draft

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

Because no annotations are supplied, the description carries the entire burden of disclosure, and it does that well: it covers append-only semantics, program-side word counting, the fact that the mirror is visible only to humans, and the nextAction protocol. It omits only smaller details such as what the write returns besides nextAction, or what happens on failure. The rich contract makes a 4 appropriate.

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 write/no-read constraint is front-loaded in the first clause, and the nextAction handling follows immediately. The redundancy is mild but defensible: the warning not to read or modify the mirror appears twice, reflecting how critical that guardrail is, though the second statement restates rather than extends the first.

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?

No output schema and no annotations exist, so the description must explain the return contract on its own, and it does: nextAction values continue and ask_user, the question and combinedText fields, and how to relay them to the user. Combined with the append-only semantics, nothing an agent needs to call the tool correctly or interpret its 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 coverage is 100%, and the schema already documents the meaning of book (matching the directory name), text (only this paragraph, no deduplication), chapter (naming the mirror file), and the single action enum value commit, which makes the baseline 3. The description echoes the paragraph-scoping rule but adds no new syntax, format, or constraint beyond what the schema states.

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 states a specific verb and resource — a mirror tool that appends the current paragraph to a per-book draft file — and explicitly marks itself as append-only, from which it follows that the AI neither reads nor modifies it. This clearly distinguishes it from the sibling tools novel_read, novel_chapters, and novel_delete_book, so an agent can tell what it does 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?

It gives explicit when-to-use rules: hand over only the paragraph being written now, then act on the returned nextAction. It also supplies negative guidance — do not read, modify, or clear the mirror, and do not decide on your own when to ask the user. The requirement to pass the question and combinedText verbatim to the user-question tool, and to stop writing until the user answers, is spelled out fully.

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