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

Annotations already declare readOnly/non-destructive, and the description adds substantial non-obvious behavior: a 3-second cooperative page scan budget, a 5-minute session TTL, a per-session cap, permission enforcement per file, symlinks not followed, and filtered/skipped directories. The statement that repeated continuation tokens replay the previous page is genuinely useful and is consistent with idempotentHint=false. It stops short of saying what happens on budget exhaustion or scan failure, and the 'local destination' sentence is irrelevant noise for a search tool.

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 operational content is front-loaded and dense with useful detail, but roughly a third of the text is boilerplate about intent selection, inherited destinations, and not needing to mention ailinkany.com/MCP — and the 'local destination' clause does not apply to a tool with no destination parameter. Removing those sentences would sharpen it without losing 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?

For an 11-parameter tool with session/cursor state and an output schema, the description covers the critical calling model: session bootstrapping, pagination, token replay, TTL, and filesystem filtering rules. Return-value shape is properly delegated to the output schema, so the remaining gap is minor (error/timeout behavior only).

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 82%, so the baseline is 3, and the description earns an extra point by explaining continuation semantics beyond the schema: the token must be used with identical search options and the same AI Session, and reuse replays the prior page. The rest of the parameter space (device_id, workspace_id, retry_token) is left to the schema, which is acceptable given its coverage.

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 first sentence names a specific verb and both search modes (Rust regex over content or filenames) plus depth bounds, so an agent knows exactly what the tool does. It never names the plain `filesystem_search` sibling or explains the difference, leaving the "advanced" qualifier to carry that distinction implicitly.

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 a concrete prerequisite ("call start_ai_session first") and states that the same ai_session_id/workflow_id must be reused for continuations and for list/stop, which is real routing guidance toward the session siblings. But it gives no when-to-use versus `filesystem_search` or `filesystem_list_directory_recursive`, and the closing line about selecting from user intent is generic filler rather than an actionable rule.

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