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

Annotations only declare readOnlyHint=false, openWorldHint=true, destructiveHint=false; the description adds substantially more: the same authorization/credit/engine checks as the primary tool, hard request-size ceilings, the requirement of an explicit user declaration (authorizationScope + liabilityAccepted), and that each batch group is tracked by its own scanId. These are constraints an agent cannot infer from the annotations alone.

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?

Opening sentence front-loads the legacy-compatibility purpose and the preferred alternative, which is exactly the ordering an agent needs. The middle sentences are dense multi-clause constructions covering limits, idempotency and authorization in a single block, so they are informative but slightly compressed rather than ideally split.

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 15-parameter, high-complexity tool with no output schema, the description covers the operational essentials: request ceilings, batching, idempotency key discipline, and the fact that results surface as per-group scanIds. It does not describe what a successful response actually contains beyond the scanId, and it defers mode/scope detail to vulkill_capabilities, leaving a small gap.

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 the baseline is 3; the description earns above that by explaining the aliasing relationships (inline_files is the legacy name for sourceFiles, targetUrl vs. single-entry dast_targets) and by stating that the two channels must not be combined or conflicted. Size limits and retry-key semantics are also restated in prose, reinforcing the schema rather than replacing it.

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 concrete verb+resource (starting a scan) and immediately frames it as a legacy compatibility entry point, explicitly routing new integrations to the sibling vulkill_start_scan. It also states which content channels (sourceFiles / inline_files) and which target channel (targetUrl / dast_targets) it accepts, so the tool is distinguishable from every sibling without opening the 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?

It gives an explicit when-to-use rule (old clients) and an explicit alternative (vulkill_start_scan for new integrations), plus batching guidance for projects exceeding 5000 files / 512 KiB and idempotency rules for retries. It goes further by ruling out local project_path and repository-URL imports, which prevents a common mis-invocation.

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