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

Annotations give readOnlyHint=false and destructiveHint=true; the description justifies the destructive flag by disclosing replacement semantics ('a new review of the same thread replaces the old one') and bounds the blast radius ('sends nothing, changes no status and records no outcome', counted in alreadyReviewedCount). It does not cover permission/auth requirements or how to undo a review, so it stops short of a 5, but it adds substantial behavior beyond the annotations.

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 purpose before branching into side effects and alternatives, and every clause carries information. It is dense with quoted examples and parenthetical asides, which slightly slows parsing, but nothing is 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?

For a 6-parameter mutation with no output schema and no annotations about permissions, the description covers purpose, effect, reversal, replacement and sibling routing. The remaining gap is the return value/confirmation (e.g. whether alreadyReviewedCount is returned), which an agent would have to discover empirically.

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 100%, so the baseline is 3, but the description adds a real cross-tool constraint: pass lastMessageAt 'exactly as li_list_conversations gave it (lastActivityAt)', mapping a confusingly named field between two tools. It also explains the thread+account keying that makes sessionId/sessionLabel interchangeable, which the schema only hints at.

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 precise verb and resource — recording that a thread was *looked at*, explicitly not an outcome about the person — and distinguishes itself from the near-identical siblings li_mark_outcome and li_mark_reply_handled by name. The scope ('without claiming an outcome about the person') is exactly the axis an agent needs to separate this from the other mark_* tools.

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?

Explicitly says when to use it (after looking at a thread and deciding something) and when not to ('use li_mark_outcome for a verdict on the person, li_mark_reply_handled to close a tracked conversation'). It also explains the downstream interaction with li_list_conversations(onlyNeedsReply) and the exact reversal condition ('a message newer than lastMessageAt brings it back'), so an agent knows both the trigger and the effect.

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