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

With destructiveHint=true and idempotentHint=false already declared, the description still adds substantive behavior: skipped steps persist in the plan, never run, and are named to the executor; the call returns every step plus its skipped state; targets, amounts, feedback sources and the selected option are preserved; prepared approval is invalidated. It also defines the step-ID addressing scheme (native, effect:0, effect:1 in the selected option's original order), which is exactly the kind of detail annotations cannot convey.

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?

Five dense, front-loaded sentences with essentially no filler; the primary action and revision constraint come first, followed by step-ID mechanics and side effects. Slightly compact to the point of being hard to parse on first read, but every sentence carries information.

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, open-world mutation with 9 params and no output schema, the description is largely sufficient: it covers return shape (every step and skipped flag), preservation guarantees, and approval invalidation. Gaps remain around revision-conflict behavior and the role of the reason field, which an agent must guess.

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 only 44% across 9 parameters, so description compensation matters, and it does explain the key semantic of skipped_steps/addressing ('effect:0, effect:1 ... in the selected option's original order') and what edits target (wording, company knowledge text) and the expected-revision constraint. It leaves reason, case and request unexplained, so some required parameters still rely on inference.

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?

States a specific verb pair (edit wording, skip steps) and a specific resource (the current saved plan at its expected revision), plus the scope of also editing proposed company knowledge text. It is clearly distinguishable from read-only siblings like summer_plans_list or summer_autonomy_plans_preview, though it never says the word 'investigations' and names no explicit alternative.

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?

Gives usable context: edits happen at an expected revision, and the description explicitly bounds scope by stating it 'does not apply knowledge or customer changes' and 'invalidates prepared approval', which routes the agent elsewhere for those actions. No sibling tool is named as the alternative, so it stops short of a 5.

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