anonymize_for_review

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

With no annotations provided, the description carries the full burden and discharges it exceptionally. It discloses exact masking transforms with an example, the reference-entry rewrite format, that prose is FLAGGED and never auto-edited, the side-effect mapping file (default name, never overwritten, must stay private), backup slot behavior, atomic validated save, and the Word-open refusal. This goes far beyond what any annotation schema would convey.

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?

Every clause earns its place: the BETA warning is front-loaded, the core purpose and reversibility come second, then specific transforms, the safety boundary, side effects, backup behavior, and refusals. The length is proportionate to the tool's complexity and contains zero filler or restatement of the name.

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 5-parameter mutation tool with no annotations, the description covers preconditions (Word-open refusal), side effects (mapping file, backups), safety (heuristic warning, never auto-editing prose, private mapping), and behavior in full. An output schema exists, so return-value explanation is not required; the reference to 'flagged-items list in the result' suffices to direct the agent to the key output.

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. author_names is explained ('Masks self-citations by the named authors'), backup is explained ('backup=False skips rotation only'), replacement is illustrated by the 'Hurd (1999)' -> 'Author (1999)' example, and mapping_path is explained with its default filename and never-overwrite guarantee. Only file_path is left to inference, and it is self-evident.

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 and resource — 'Anonymize a manuscript for double-blind peer review, reversibly' — with a clear distinguishing property (reversibility) that separates it from irreversible siblings like redact_text, and names its reverse counterpart deanonymize_document. The scope (self-citations, reference entries, metadata) is explicit enough that an agent cannot confuse it with other document tools.

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?

Provides clear context: this is for double-blind peer review where reversible anonymization is needed, pairs with deanonymize_document, and warns the agent to review the flagged-items list rather than trust output silently (BETA heuristic). It does not explicitly enumerate alternatives or state when NOT to use it, but the conditions for correct use and the failure precondition ('Refuses documents open in Word') are present.

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