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

The description discloses far more than the readOnlyHint/destructiveHint annotations already provide: it confirms no allocation is computed or recomputed, pins ordering to period/version/run id, explains the page_token pinning contract and source_changed restart, clarifies that current_only means version-freshness not source-freshness, and states that known_stale and not_assessed are warnings only with raw stale reasons never returned. All statements are consistent with the annotations, so there is no contradiction.

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 dense (roughly 130 words) but front-loaded with purpose, and every clause carries distinct operational information — ordering, pagination, warning semantics, money format, and response omissions. It loses a point for length and for the cryptic cross-reference "Choose one allocation_dimension in a detail read," which is not resolvable from this tool's own schema.

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 full burden of explaining result behavior, and it covers the risky parts: result grouping and ordering, pagination restart, staleness warning semantics, and what is deliberately not returned. The only gaps are the unresolved allocation_dimension cross-reference and the absence of any concrete response-shape hint, which are minor for a read-only discovery tool.

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 all five parameters are already documented, including page_token's restart-on-source_changed behavior and current_only's "not a freshness claim" caveat. The description reinforces these semantics (filtered-source pinning, current_only vs source freshness) but adds little that the schema does not already state, so the baseline of 3 applies.

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 description opens with a specific verb and resource: "Discover persisted allocation runs and their separate cost-centre and profit-centre reconciliations." It also draws an explicit boundary against transformation tools with "This does not generate or recompute allocation," which distinguishes it from compute/generate siblings. The "MANAGEMENT data class" qualifier reinforces its discovery role.

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 clear operational context: detail reads should "Choose one allocation_dimension," reconciliations must not be summed ("do not add them together"), and the current_only and warning semantics shape how results should be interpreted. However, it never names a specific alternative sibling tool or states an explicit when-not-to-use condition, so routing guidance is implied rather than stated.

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