undo

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

No annotations are provided, so the description carries the full behavioral burden. It discloses that a change is reverted and a preview is returned, but not whether the revert is destructive or reversible, whether repeated calls step further back, or how failure with an empty history is signalled. For a state-mutating tool with zero annotation coverage this is a significant gap.

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?

A single short sentence with the action front-loaded and the return value appended. No wasted wording, no redundancy with the name.

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 zero-parameter tool with no output schema, saying it returns the preview is genuinely useful. However, the missing details on undo depth, history preconditions, and failure behaviour leave real gaps for an agent that must decide when this call is safe.

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 tool takes no parameters, which is the baseline-4 case. The description correctly implies nothing needs supplying and that the target is implicit (the last change).

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 gives a specific verb ('Revert') and a specific resource ('the last recipe change'), and adds that it returns the preview. An agent can distinguish this from siblings like develop, frame, or get_state. The only weakness is that 'recipe change' is domain jargon that doesn't pin down which edits count as undoable.

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 guidance on when to call undo instead of inspecting state with get_state, or on whether undo is single-level or repeatable. Nothing says what happens when there is no prior change to revert, so the agent must guess at preconditions.

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