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

Annotations already declare a non-readonly, non-idempotent, open-world, non-destructive operation, and the description does not contradict them. It adds real context beyond annotations — that it returns a resumable run, that existing directions are counted/reused, and that it is gated on user permission and a deliberate choice before building.

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?

Four compact sentences with the purpose front-loaded and no filler. A couple of clauses are terse/telegraphic ('follow the user's permission'), but each sentence carries actionable 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?

With no output schema, the description takes responsibility for the return value ('Returns a run to resume with get_concepts') and the surrounding workflow, which is what an agent needs to chain calls. It stops short of describing what the run contains or failure behavior, but is largely complete for invocation.

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 moderate (67%). The schema itself already documents count (up to 12, default 2, reuses existing) and direction (plain-English steer) well, and the description adds little parameter-level detail beyond 'for an existing audit' mapping to audit_id. Baseline 3 is appropriate since the schema does most of the work.

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+resource+scope: 'Generate visual design directions for an existing audit.' It also distinguishes itself from siblings by naming check_balance (coverage) and get_concepts (retrieve the run), so an agent can tell generation apart from retrieval and building.

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?

Explicit workflow rules are given: generate 'when requested', use check_balance for page coverage, respect the user's permission, show existing images before asking to choose, and never start a build without their choice. Alternatives (check_balance, build_page implicitly) and preconditions are stated rather than implied.

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