Tower preview

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

Annotations declare destructiveHint=true, readOnlyHint=false and openWorldHint=true, yet the description calls the operation a 'preview' and says nothing about side effects, scope of impact, or authorization requirements. The 'preview' framing sets up a read-only expectation that the annotations contradict, which is a serious omission even if not an explicit false claim. The one real addition is the evidence-handling rule (untrusted data, cite record IDs, preserve uncertainty), which is useful but concerns output handling rather than what the tool does to state.

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 short sentences with no filler and the qualifier 'Authenticated' front-loaded. However, the brevity comes at the cost of substance rather than from tight editing; the sentences do not map to purpose, parameters, or effects.

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 tool with three undocumented parameters and no output schema, the description should cover what is previewed, what state can change, and what the response contains. None of that is present; only an evidence-hygiene note is supplied.

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% and there are three parameters (incident_id, dependency, expires_in_seconds), none of which appear in the description. The meaning of 'dependency' and the effect of 'expires_in_seconds' — critical for a destructive operation — are left entirely undefined.

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 largely restates the title ('Tower preview' -> 'Authenticated Control Tower preview'). It never says what is being previewed (a runbook execution? an incident remediation plan?) or what the tool produces, so an agent cannot distinguish it from siblings like tower.runbook_preview or tower.investigate.

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?

No when-to-use guidance is given, and no sibling alternative is named despite many plausible overlaps (tower.evidence, tower.investigate, tower.runbook_preview). The single boundary hint — 'Human approval uses Tower HTTP/CLI separately' — tells the agent this is not the approval step, but not when to pick this tool.

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