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

Annotations already declare destructiveHint=true and idempotentHint=true, so the safety profile is known; the description still adds real value by scoping the destruction ('Autopilot, Off/Pause/ask unchanged') and prescribing a fresh UUID clientRequestId. It does not cover failure modes such as revision conflicts or what happens to dependent artifacts that referenced the cleared plan.

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?

Extremely compact and front-loaded, with zero filler, but the telegraphic fragment style borders on cryptic and forces the agent to infer sentence structure between the clauses.

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 prerequisites, idempotency, and scope of destruction for a destructive mutation tool with no output schema and no annotation text. It leaves unstated the return value and optimistic-concurrency error behavior, which matter for a destructive clear.

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%, so the description carries the whole burden, and it does map all three parameters to their sources. However, the wording is shorthand ('planRevision', 'control.expectedUpdatedAt') and does not spell out the field-to-parameter mapping or UUID format constraints beyond 'new UUID'.

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 (erase) and resource (saved plan), plus an explicit scope boundary: 'only' the plan, while Autopilot/Off/Pause/ask are unchanged. That boundary disambiguates it from adjacent plan siblings, though no sibling is named directly the way soclo_agreed_plan_read is used as a prerequisite.

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?

Tells the agent what to do first ('planRevision, control.expectedUpdatedAt from soclo_agreed_plan_read') and where to go when the request is unclear ('Unknown: soclo_agreed_plan_intent'). Clear routing, but no explicit statement of when not to clear or what to use instead for a simple edit.

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