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

The description discloses rich behavior beyond the sparse annotations: fire-and-forget semantics, server-side deduplication with 'deduped: true' responses, drift detection via assistant_message_sha256 ('drift: true'), and safe re-send behavior. While idempotentHint=false is conservatively set, the description precisely documents the actual dedup guarantees (content-based within a short window, turn_id-based forever), which supplements rather than contradicts the annotation.

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 every sentence earns its place — the most critical behavioral constraint (non-blocking, reply-first) is front-loaded, and each paragraph covers exactly one parameter or policy. Given 0% schema coverage and 7 parameters, this density is appropriate rather than verbose.

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-value documentation is not required, and the description covers everything else an agent needs: call ordering, retry and catch-up semantics, parameter semantics for all non-obvious fields, idempotency behavior, and the integrity-checking mechanism. No operational gaps remain for a tool of this complexity.

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 bears the full burden — and it delivers. chat_id ('omit or null to start a new conversation; otherwise reuse the id returned by a prior save_turn'), source (with concrete examples), attachment_context (same summary as get_context), turn_id (retry never double-saves), and assistant_message_sha256 (integrity check, not a gate) are all given meaning the bare schema lacks. It also clarifies that assistant_message is 'the reply as actually sent, not a draft.'

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 opening sentence — 'Persist this exchange' — names a specific verb (persist/save) and resource (the conversation turn), making the tool's function instantly unambiguous. It is clearly differentiated from the sibling list, which is dominated by retrieval tools (search_keyword, recent_messages, get_context), and this is the only persistence/write tool for conversation history.

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?

The description gives explicit sequencing guidance: 'Send your reply to the user FIRST, then call this after (or alongside) it; do not block or delay the user-visible reply.' It also specifies failure-handling policy (retry once if the environment allows, otherwise catch up on the NEXT turn) and cross-tool consistency with get_context for attachment summaries. This is textbook when-to-use guidance.

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