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

Annotations already declare readOnlyHint=false, destructiveHint=true and idempotentHint=true, so the higher-order safety profile is covered. The description adds real behavior the annotations can't express: media is preserved rather than replaced, and text may only be empty while media exists. It stops short of saying whether omitted optional fields (content, sort_order, active) are cleared or preserved — a material question for a full-replace operation with destructiveHint=true.

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?

Three short, front-loaded sentences with no filler; scope is stated before the media exception. The opening line 'Replace a saved reply' duplicates the title verbatim and earns little, which is the only real waste.

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 7-parameter destructive replace with idempotent semantics and no output schema, the description covers intent and the media/content rules well. The significant omission is replacement semantics for the three optional fields: an agent cannot tell whether omitting content or sort_order wipes them or leaves them intact, which matters given destructiveHint=true.

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 86%, so the baseline is 3. The description maps the prose fields to parameters (title, shortcut, text/content, order/sort_order, active) and reinforces the content-vs-media rule already in the schema. It adds no new syntax or value constraints beyond what the schema descriptions provide.

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?

Goes past the title restatement to enumerate exactly what is replaced — title, shortcut, text and order — plus the `active` toggle. It also names the sibling tools that handle media (importSavedReplyMedia/removeSavedReplyMedia), which separates this tool's scope from adjacent ones. It does not explicitly contrast with create_saved_reply, but the verb 'replace' plus the required saved_reply_id makes the distinction unambiguous.

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?

Explicitly routes media changes to importSavedReplyMedia or removeSavedReplyMedia rather than this tool, with the condition 'media stays'. The content constraint ('empty only while it has media') tells the agent when an empty value is legal. No when-not guidance for the tool as a whole (e.g., use create_saved_reply when the reply doesn't exist yet), which keeps it below a 5.

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