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

Annotations already flag destructiveHint=true and readOnlyHint=false, yet the description goes far beyond them: it names exactly what is restored and removed, the refusal conditions (different model, model edited since apply), that force overwrites those edits, resumability semantics, and that the website apply runs as a background job. This is the kind of context annotations cannot express.

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?

Front-loads the core action and effect, then layers refusal conditions, resumability, default run, and preview/background behavior. Dense but nearly every sentence carries distinct information; the resumability and background-job clauses could be trimmed slightly without loss of safety-critical meaning.

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, resumable mutation tool with no output schema, the description covers outcome (model returns exactly), failure reporting, retry behavior, and dry-run default. An agent has everything needed to decide whether and how to call it, including the destructive override path.

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 67% and leaves apply undocumented, but the description compensates fully: 'Previews unless apply: true' explains apply, force:true is described as undoing anyway while overwriting edits, and runId's default (latest not-yet-undone run) is spelled out. Each parameter's behavioral effect is clear, not just its name.

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 (undoes) and resource (a bridge_reshape ledger run), and describes the exact effect: rows overwritten are written back, rows created are removed. It is clearly distinguishable from bridge_reshape (which applies) and bridge_reshape_read (which reads) without inspecting any schema.

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 concrete when-to-use context: default targets the latest not-undone run, apply:true is needed to execute, force:true overrides the 'model edited since apply' exclusion, and re-calling retries only failed steps. It does not explicitly name bridge_reshape_read or bridge_reshape as the alternative to pick for non-undo needs, but the exclusions and prerequisites are unusually complete.

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