Undo

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

Goes far beyond the annotations: EXPERIMENTAL status, undocumented QE DOM paths, project-wide undo history, step-by-step position check with a stated blind spot ('matching position alone does not prove which action is on top'), persistence of the marker barrier across reloads, and enumerated limits (UXP, manual UI, engine resets). The destructiveHint=false annotation is not contradicted, though the warning that Undo 'would reverse an earlier action' is exactly the risk an agent must weigh.

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 dense paragraph with the critical EXPERIMENTAL warning and safety-guard behavior front-loaded, but verbose and repetitive: the caveat that a matching index proves position only is stated twice. Much of the length is earned by the tool's complexity, but it could be tightened.

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 mutant, experimental QE-backed tool with a required safety guard, the description covers prerequisites, failure modes (refused step, marker barrier), permission/acknowledgment semantics, and known blind spots. An output schema exists, so return values need not be described.

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 100%, so baseline is 3, but the description adds real meaning: it explains that the undoSteps value from a prior call should be passed as count to reverse exactly that call, and that expected_undo_stack_index is a positional guard rather than proof of the top action. That connects the parameters to the workflow rather than restating their types.

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 names a specific verb and resource ('Undo the most recent Premiere project action(s) through QE') and pins down the mechanism (PS undo-stack position verified via undoStackIndex). It implicitly scopes itself away from untracked actions, but never names the adjacent siblings redo or multiple_undo, so an agent must infer the boundary itself.

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?

Strong when-to-use guidance: only actions Premiere records are undoable, pass the undoSteps a QE edit returned as count, and acknowledge_untracked_markers is required to pass a marker receipt boundary. Clear conditions, but no explicit routing against the sibling undo-family tools.

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