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

Annotations (readOnly, idempotent, non-destructive, closed-world) cover the safety profile, and the description adds genuinely new behavior: server-side vs browser delegation semantics for `auto`/`server`/`browser`, gpu:true always delegating, transient non-retained input processing, and an AP2 artifact with execution_hash for chain provenance. That is above the annotation baseline, though it omits rate limits and what the returned artifact actually contains.

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 purpose is buried behind infrastructure boilerplate, and the trailing FV-status sentence plus a 64-character hash URL consumes a large share of the text without helping an agent decide whether to call the tool. The sentence 'Inputs are processed transiently...' relabels the same point made twice ('not stored, logged, or retained').

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 values need not be described, and compute/verification behavior is reasonably covered. However, for a nested-object analytics tool the domain contract is missing: what `policy_parameters` should contain is punted to an external manifest, and nothing explains what pricing quantity the node produces.

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 100%, so the schema already documents compute, parent_hashes, parent_tool_ids and policy_parameters. The description only restates the compute-mode semantics already spelled out in the schema, and defers policy_parameters field names to 'the tool's manifest' rather than adding anything — baseline 3.

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 title/description identifies the resource ('Embedded Insurance Pricing Modeller') but the body never states what is actually computed — premium, commission, loss ratio, coverage tranche — and just labels itself 'OpenChainGraph compute node (analytics_mandate)'. A reader learns the plumbing (compute modes, artifact export) rather than the analytical purpose, so it is only separable from the ~hundreds of sibling `compute_*` nodes by name, not definition.

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?

There is one real constraint — 'Use synthetic or anonymised inputs only' — and an implied choice between compute modes. But there is no guidance on when this pricing model applies versus alternatives (e.g. other pricing/insurance siblings), no prerequisites, and no indication of what the tool is *not* for.

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