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

Annotations only say non-read-only, destructive, non-idempotent; the description adds far more: upsert semantics ('creates the criterion if missing, else updates'), the confirm=false preview path, and Google's measured rejection of partial adjustments on Demand Gen with a friendly error reply. Minor tension: 'creates if missing else updates' reads as an upsert rather than strictly non-idempotent, but this is a nuance, not a contradiction.

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-loaded with the core operation, then structured with bullets; almost every line carries unique information. Slightly verbose in places (the percentage examples and the recipe partially restate each other), but no real filler.

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?

An output schema exists so return values need not be explained; the description instead covers everything an agent needs: upsert behavior, preview mode, device syntax, modifier range, and channel-specific validation quirks.

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 0%, so the description carries the full burden and does: it enumerates device values (MOBILE|DESKTOP|TABLET|CONNECTED_TV), comma-separated syntax, and fully explains bid_modifier_pct semantics (-100 exclude, 0 neutral/reset, positive/negative percentages, range -100..+900). It even documents confirm's preview behavior.

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 and resource ('Set a campaign's per-DEVICE bid adjustment') and explicitly positions itself as 'the WRITE side of the device criteria that get_campaign_targeting reads,' cleanly separating it from the read sibling.

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?

Gives concrete when-to-use recipes (mobile-only = device 'DESKTOP,TABLET' with -100), names the CONNECTED_TV exclusion case for Demand Gen/Video, and even states the channel-specific constraint that partial adjustments are rejected there. An agent knows exactly when and how to call it.

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