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

Annotations already declare destructive/readOnly/idempotent/openWorld, but the description goes further: patch carry-forward vs. explicit-empty-list clearing (the real destructive semantic), CAS concurrency via expectedRevision, idempotency enforced by a stable clientRequestId, required manage:settings permission, and the agent_config_missing failure path. It stops short of describing the response payload, but that is a minor gap against this much added context.

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 sentence carries information and the safety-critical sequencing comes early, but the opening sentence is a dense run-on of conditional scope forms that front-loads syntax before stating the tool's action. Slightly compressed for an agent parsing under time pressure.

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?

Covers the full burden for a nested, mutation, 0%-coverage, no-output-schema tool: prerequisites, ordering, revision concurrency, idempotency, permission, and an error path. Nothing an agent needs to call it correctly is 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 description coverage is 0% and directionScope is an untyped empty schema, so the description is the only source of meaning — and it delivers: directionScope's three accepted shapes (line/actionId, range/startDate/endDate, ongoing/startDate), expectedRevision as the value from read, clientRequestId as a stable UUID, confirm as the two-step gate, and the semantics of omitting vs. clearing direction/templates/targets.

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+resource ('Saves only creative_preferences through the app's CAS RPC') and explicitly bounds scope by naming what it never does (enroll, enable, schedule, pay for Autopilot), which is exactly the confusion it would share with the autopilot_* siblings. An agent can tell it apart from soclo_creative_preferences_read and soclo_autopilot_save without opening schemas.

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?

Explicit sequencing: read soclo_creative_preferences_read first, pass the exact expectedRevision, call once without confirm to review, then confirm the same intended patch. It also names the alternative (the read tool) and states the when-unclear rule ('Ask one post, day, several days, week, month or ongoing when unclear; never assume').

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