vm_checkpoint

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

With no annotations, the description carries the full behavioral burden, and it omits key traits: whether this requires an active session (vm_connect), whether it overwrites or appends to prior checkpoints, whether it is idempotent, and what happens on failure. The "no secrets or full transcript" exclusion is a genuinely useful safety constraint, which keeps this above the floor.

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 short sentences, front-loaded with the action and immediately followed by the constraints. No filler, though the field list reads as telegraphic and could be ordered to mark what is required vs. optional more clearly.

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 mutation tool with a nested free-form object, no annotations, and no output schema, the description should explain the checkpoint lifecycle, session prerequisites, and the identity parameters. Instead it covers only the body payload and one content restriction, leaving an agent guessing about tenancy and overwrite behavior.

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 coverage is 0% and the body parameter is a free-form object with additionalProperties and no declared properties, so the description's enumeration of body fields (objective, verified facts, uncertainty, next_step, optional scope/evidence) is real added value. However, the required vm and owner_thread_id parameters get no explanation at all, leaving half the surface undocumented.

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?

"Save" plus the enumerated payload fields (objective, verified facts, uncertainty, next_step, scope/evidence) indicate a state-persisting tool, so the action is inferable. However, it never defines what a "checkpoint" is relative to siblings like vm_pin, vm_observe, vm_experience, or vm_propose_update, so an agent cannot tell it apart from other persistence tools by reading this alone.

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 when-to-use guidance, no trigger condition, and no named alternative among the 22 siblings. The only guidance is a content restriction (no secrets / no full transcript), which constrains what to write rather than when or why to call it.

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