undo

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

With no annotations provided, the description carries the full burden of disclosing behavioral traits, and it excels: it reveals three critical non-obvious behaviors—atomicity on the prior execute call, full wiping of mixed-content scripts, and a preference for delete-loop cleanup as an alternative strategy. For an `undo` tool whose destructive reach and edge cases are non-obvious, this is exactly the kind of cautionary context agents need. The description reveals consequential, non-obvious side effects that would otherwise be discovered only through failure.

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?

Three sentences, front-loaded with the core purpose followed by essential warnings, with zero filler words. The structure parallels the canonical training example of how concise, high-value information should be presented.

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?

Given a minimal 1-parameter schema and no output schema, the description covers all the critical surface area: purpose, count semantics, and destructive behaviors. There is slight ambiguity around how `count` interacts with the atomic undo of a prior execute (does count>1 chain multiple undos?), but for the tool's complexity, this is a minor gap.

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 must explain the `count` parameter, and it does: "Undo the last `count` actions" directly ties the parameter to its semantics. While it could add caveats (e.g., behavior when count exceeds history or is zero), for a single optional parameter with a default, this is adequate, valuable coverage.

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 opens with a clear verb+object construction, "Undo the last `count` actions," which unambiguously states what the tool does. It goes beyond a tautology by scoping the operation to a count of actions, and it stands distinct from its `redo` sibling by definition. The purpose is immediately, unmistakably clear.

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?

The description provides explicit behavioral cautions with the WARNING and directs users to "Prefer delete-loop cleanup when possible," which is a clear alternative-approach recommendation. It doesn't name a specific sibling tool, but it tells agents when NOT to rely on it (atomicity limits, mixed-content wipes), which meaningfully informs tool selection.

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