recovery_restore

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

Annotations already declare destructiveHint=true and readOnlyHint=false, so the risk profile is known. The description adds genuinely new behavior beyond that: the commit ordering ('Clears the fence only after full verification') and the reporting contract ('an alternative target is reported as recovery, never as rollback'), which tells the agent how to interpret the outcome. It still does not say concretely what pre-write work is discarded or whether the operation is reversible, keeping it short of a 5.

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?

Two dense sentences with the default behavior front-loaded and the conditional alternative immediately after; no filler or restatement of the name. The density tips into opacity in places ('the fence'), but nothing is wasted.

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 destructive recovery operation with no output schema, the description covers the post-condition (verified commit) and the reporting semantics, which is the essential part. It leaves gaps on prerequisites (whether a recovery or transaction must be open first), where checkpoint IDs come from, and what is irreversibly lost, which matters for a destructive tool in a cluster of similarly named restore siblings.

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 optional parameter has 0% schema description coverage, so the description carries the burden. It does add meaning not in the schema: omitting checkpoint_id defaults to the pre-write model, while supplying one requires a checksummed managed checkpoint that predates the failure and lives in the same saved scene/topology. It does not state how to obtain a valid checkpoint_id (presumably checkpoint_list/create/status), leaving that link implicit.

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 states a specific verb ('Recover') and target ('recorded pre-write model', or a checksummed managed checkpoint), so the operation is identifiable. However, it is delivered in undefined domain jargon ('fence', 'pre-write model', 'saved scene/topology') and never distinguishes itself from the many siblings that also restore state (checkpoint_restore, recovery_open, pose_restore, subtree_restore). An agent can guess what it does but not confidently why it is the right tool.

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?

There is an implied post-failure context and one conditional branch ('or explicitly choose a checksummed managed checkpoint predating the failure'), which gives some sense of when the alternative parameter path applies. But there is no explicit when-to-use, no when-not-to-use, and no reference to the near-identical siblings (checkpoint_restore, recovery_open, recovery_verify, recovery_status) that an agent must choose between.

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