checkpoint_create

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

Annotations already declare destructiveHint=true and readOnlyHint=false, so the agent knows this mutates state; the description adds only that the result is "durable" and "bounded." It never explains what gets destroyed or overwritten, what "bounded" means for existing checkpoints (eviction of the oldest?), or that a checkpoint likely consumes a slot needing checkpoint_release. For a destructive mutation these are the most useful things the description could have said.

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?

One front-loaded sentence with a semicolon-delimited scope caveat; nothing is redundant. "Bounded durable checkpoint" is slightly jargon-dense but each phrase carries content.

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, non-read-only mutation with no output schema and no documented lifecycle, the description leaves the agent guessing about capacity limits, whether prior checkpoints are evicted, and how this pairs with checkpoint_restore/release. The scope-exclusion clause is a useful start but not enough for a tool with this risk profile.

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?

There is a single optional parameter, label, with 0% schema description coverage, so the description ideally would explain naming/identity semantics. It does not mention label at all. With only one optional param and no nesting the gap is modest, but it is a real one rather than a case of the schema doing the work.

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?

"Capture every stored model channel and track into a bounded durable checkpoint" gives a specific verb (capture) and resource (animation channels/tracks -> checkpoint), which distinguishes it from snapshot-style siblings like pose_snapshot or subtree_snapshot. The negative clause "no native save-as or asset copying" further bounds scope. It stops short of naming the checkpoint_* siblings it belongs to, so it is clear but not fully differentiated.

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 no statement of when to create a checkpoint versus using checkpoint_status, checkpoint_restore, or the pose_snapshot family, and no prerequisite or timing guidance. The only guidance is a scope exclusion (it does not save files or copy assets), which is not the same as when-to-use advice.

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