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

Annotations already declare readOnlyHint=true, destructiveHint=false, and openWorldHint=false, so the safety profile is covered; the description adds real behavioral context on top: PII facts are masked as bullets, maskedKeys lists the masked fields, and unmasking is console-only and audited. It also explains the null-layer semantics with three named causes (COMPOSITE_EXECUTION, NOT_LIVE_VERSION, NOT_RECORDED), which is exactly the kind of partial-result disclosure an agent needs. It stops short of error behavior or how to obtain a traceId, so not a 5.

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 core purpose is front-loaded in the first clause, which is good, but the remainder is one long run-on sentence with deeply nested parentheses and compressed jargon like 'deterministic why per rule'. The three unavailableReason cases do earn their place for interpretation, yet the overall density makes it harder to scan than necessary for a one-parameter read tool.

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, the description carries the return-value burden and does so for the main lineage components plus the null-layer cases, which is strong for a read-only single-parameter tool. The main remaining gap is that it never says where a traceId comes from (e.g., from a replay or log result), which an agent needs in order to actually invoke it.

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?

There is a single parameter and schema description coverage is 100%, with 'Trace ID of the execution' documented inline, so the baseline of 3 applies. The description's phrase 'a single decision' reinforces the one-trace-per-call scoping but adds no format, sourcing, or lookup guidance for traceId beyond the schema.

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 clause names a specific verb and resource ('Get the lineage of a single decision') and then enumerates exactly what that lineage contains: the decision, per-rule deterministic why, input facts, the responsibility chain, and the rule snapshot fingerprint. That level of specificity lets an agent distinguish it from adjacent read tools such as lexq_replay_decision or lexq_history_get without opening the schema.

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 implied by the content list (reach for this when you need provenance/audit detail for one decision) and the unavailableReason semantics hint at when the answer will be partial. However, it never states a when-to-use condition, any exclusion, or which sibling to prefer for related questions like replaying or browsing history. The agent must infer selection from the return contents alone.

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