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

Annotations only declare read-only/idempotent, so the description carries real weight and does: it discloses data provenance (Silicon Analysts estimates, capacity_basis 'sa_estimate', companies do not disclose per-fab capacity), that availability_status is OUR tier not the company's statement, the period_relation vs is_projection pitfall, the no-carry-forward rule for lapsed series, meta.corrections, and differing access per tier (anonymous/free preview vs Pro full series). These are exactly the non-obvious behaviors 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?

Purpose and mode semantics are front-loaded and every sentence carries substantive caveats, but the opening block is a dense wall of parenthetical detail that would benefit from tighter grouping. No sentence is pure filler, so it stops short of a 5.

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 7-parameter, no-output-schema analytical tool this is complete: provenance and quoting rules, tier-based access differences, caveats about stored flags versus derived relations, corrections metadata, citation format, and the sibling routing are all present.

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%, so the description must compensate, and it names all seven filters with added meaning: as_of is explicitly a latest-state cutoff, series selects full dated history (with its gating), and limit is defined in row terms (fab·class rows in latest state, newest readings when series=true). That clarifies semantics beyond the bare JSON types.

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: per-fab, per-tech-node-class capacity estimates built from fabs + fab_capacity_snapshots time-series, with default latest-state vs series=true modes spelled out. It explicitly distinguishes itself from get_bottlenecks, get_foundry_allocation and get_wafer_pricing, so an agent can route correctly without opening a 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?

Dedicated 'USE THIS for' examples and a 'DO NOT USE for' block that names the alternative tool for each excluded case (get_bottlenecks for evidence-graded supply questions, get_foundry_allocation for allocation/lead-time per node, get_wafer_pricing/foundry endpoints for annual wafer starts). This is the explicit when/when-not/alternative pattern at its strongest.

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