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

Annotations already declare readOnly, idempotent, non-destructive, closed-world; the description still adds genuine context beyond them — no provider calls or credits are consumed, provider receipts remain unverified, and a saved draft is neither scheduled nor published. It does not cover auth or ownership failure modes, but the core behavioral boundaries are disclosed.

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?

Five short sentences, leading with what is read and following with recovery, safety boundaries, and next step — well front-loaded and free of filler. The first sentence is a dense comma-separated list that takes a reread, which keeps it from a 5.

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?

An output schema exists, and the description still conveys the shape of what comes back (plan, materials, approval link, phase, evidence) plus the trust caveats (unverified receipts, unscheduled draft) and the required follow-up action. For a two-parameter read tool this leaves nothing an agent needs 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 both parameters are documented, so the baseline is 3. The description adds meaning the schema cannot express by stating that the two identifiers are alternatives ('either attempt_id or the original preparation_key'), which is important because the schema marks neither as required and does not encode a oneOf constraint.

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: 'Read one originating-key directory plan' and enumerates the payload (public materials, native approval link, client-reported phase, bounded independent evidence). That is far more specific than a title restatement and separates it from the many other get_* siblings, though it does not name those siblings explicitly.

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 a clear conditional use case ('Supply either attempt_id or the original preparation_key to recover a lost response') and a sequencing rule ('follow with webanto_submit_directory_handoff only after native approval'). What is missing is explicit exclusion guidance versus adjacent read tools, so the agent must infer that this is the only tool covering directory handoffs.

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