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

Beyond the annotations (non-destructive, idempotent), the description discloses cost ("Free"), a deliberately skipped side effect (the quality check is not re-run), validation behavior (a change introducing a lint error is refused), and the response payload ("previous: the old values, to send back to undo"). That is exactly the undo/validation/failure information an agent needs before mutating a live template.

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?

Dense but tight: the access gate, the editable field list, the partial-update rule, the skipped QA step, the undo payload, and the lint refusal each appear once and in a usable order. The opening fragment "Fetch Hive staff only" is telegraphic enough to read as a stray token before its meaning lands, which slightly hurts front-loading.

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 13-parameter mutation tool with no output schema, the description covers the mutation semantics, the undo path, the validation failure mode, and the missing-QA caveat that an agent would otherwise get wrong. The residual gap is one unmentioned parameter (`max_tool_calls`) and the absence of any value-range guidance, which matter less because the schema encodes those constraints.

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?

With schema description coverage at only 15%, the description carries most of the semantic load by mapping plain-language field names (instructions, model, temperature, reasoning effort, token limits, example message, name, description, how it works, good for) onto the cryptic properties such as `instruction_prompt` and `max_thinking_token`. It leaves `max_tool_calls` unnamed and does not convey the numeric bounds or string length limits present in the schema.

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 names a specific verb and resource ("Change an agent template already in the library") and enumerates the editable surface, including the lifecycle states it applies to (draft, pending review or live). That clearly separates it from idea-level or step-level edit tools, though it never names the closely named `update_agent` sibling to draw the boundary 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?

"Pass only what changes" is an explicit partial-update rule, and the description states a precondition (Hive staff only) plus a required follow-up action (re-run QA in fetch-admin afterwards). It gives clear context but never states when to prefer a different editing tool or when a change should instead be a new template.

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