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

No annotations are provided, so the description carries the full burden and meets it: it discloses cost/latency (two live windows per call), data-quality caveats (impression-weighted positions, sparse data at depth, treat absolutes as directional), and exact pagination semantics including the truncated flag and the consequence of ignoring it. This is behavioral context an agent cannot get from the schema.

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?

The description is long but front-loaded with the purpose and then broken into clearly labeled POSITION and PAGING blocks, so an agent can skim to the relevant section. Nearly every sentence adds a distinct constraint, though the prose is denser than it needs to be.

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 and no annotations, the description compensates by describing the return shape (returned, total_available, truncated, next_offset per list) and the diagnosis categories, plus the interpretation caveats for positions and truncation. An agent has everything needed to call it and to read the result correctly.

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 75% and the schema already documents days, limit and offset, so the baseline is 3. The description adds real value beyond it: it explains that limit is ordered biggest-loss-first, that the response's total is the number an agent needs for a 'how widespread' question, and that offset should be taken from pagination.<list>.next_offset. Only site_id remains unexplained.

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 and a precise scope: pages that lost clicks between the current window and the one immediately before, with a named diagnosis (ranking loss, CTR collapse, falling demand, mixed). It distinguishes itself from the cache-reading family by cost ('fetches two live windows per call, so it is slower'), but never names a specific sibling such as Live_Content_Decay or Live_CTR_Opportunities, so overlap with those live tools is left to inference.

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?

Usage is only implied: the latency disclosure ('slower than the tools that read the synced cache') tells an agent this is the fresh-data option, and the paging instructions govern correct invocation. There is no explicit 'use this when X, prefer Y when Z' routing against the several other Live_* siblings, so guidance is adequate but inferential.

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