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

Annotations already establish readOnly/idempotent/non-destructive, and the description goes well beyond them: expiry window (7 days default, 1-30 configurable, surfaced as ttl_days), the bounded store with default retention of 50 points / 100 MB dropping oldest first, the non-obvious fact that listing does not prune but the next recorded point will, the 'undo disabled on site' branch, and the per-connection reachability caveat via the reversible flag. This is exactly the operational context annotations cannot carry.

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?

Front-loaded with purpose then the return shape, and every later sentence carries a distinct operational fact. It is long, and 'Only the account's own undo points are visible' restates the scope already given in the opening clause — one sentence that doesn't fully earn its place.

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?

With no output schema, the description correctly enumerates the returned fields (token, tool, kind, summary, items, bytes, created_at, expires_at, reversible, count, ttl_days) and explains the one ambiguous field, reversible. Combined with the retention and expiry rules, an agent has everything needed to interpret results without a follow-up call.

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?

The single `limit` parameter is fully documented in the schema (range, clamping, zero-as-default, default 20), so with 100% coverage the schema does the heavy lifting and the description adds nothing about it. Baseline 3 is appropriate; no parameter detail is missing, but none is contributed either.

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?

Specific verb and resource ('List the account's own undo points') immediately glossed as 'the snapshots that destructive tools took before they changed something', which tells an agent exactly what class of object this returns. It is cleanly separable from the write-side siblings wp_undo and wp_undo_discard, which are named later.

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 closing routing sentence ('wp_undo restores a point; wp_undo_discard drops one') makes the alternative tools and their roles explicit, and 'Listing changes nothing' clarifies that this is the safe inspection step before restoring. It stops short of an explicit 'do not use this to restore' phrasing, but the intent is unambiguous.

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