Run the design deviance scan

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

Adds substantial context beyond annotations: a background scan capping at 1,500 files, the 429 RATE_LIMITED rate-limit ceiling, and the exact result shape ({ refresh, run } with running status or null). Crucially it reconciles the readOnlyHint=false annotation with 'it records a scan; nothing in the repo changes', so the agent understands the write is metadata-only.

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 action and scan target, then packs behavioral facts efficiently into two dense sentences. It is information-rich rather than padded, though the run-shape detail could be tightened slightly.

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?

With one param, no output schema, and high schema coverage, the description supplies everything an agent needs: what it scans, the polling pattern via the sibling, rate limits, and the null-result fallback. Nothing required to call it correctly is missing.

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 100% and the single projectId parameter is fully documented in the schema, including the default. The description only repeats that projectId is optional, adding no new syntax or format meaning, so the baseline 3 applies.

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 ('start a new deviance scan') plus resource ('design tokens and repo') and enumerates exactly what it detects (off-token colours, fonts, spacing, radius, contrast below AA, raw interactive elements). It is clearly distinguishable from the sibling get_design_deviance, which it frames as the polling counterpart.

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?

Names the alternative (get_design_deviance) with the condition that selects it (poll until running clears) and states the trigger constraints (one scan per project per 5 minutes, run null when no tokens). It stops short of saying when a caller should run a scan at all versus just reading tokens, but the routing guidance is solid.

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