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

The description substantially exceeds the annotations by disclosing deterministic execution, server-side Cloudflare Workers behavior, browser-delegation semantics, transient input processing with no retention, synthetic/anonymised input constraints, AP2 artifact export, and execution_hash provenance. This gives an agent a strong operational model beyond the readOnly/idempotent/destructive hints.

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 information-dense and front-loaded with purpose and compute semantics, but it is quite long and includes auxiliary material such as the Open URL and FV-status snapshot details that are not necessary for correct invocation. A more scannable structure with separate operational, data-handling, and provenance sections would improve readability.

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 tool with nested parameters and no output schema, the description covers execution location, compute-mode selection, data retention, output artifact shape, and downstream consumers. It does not describe the exact response format or resolve the policy_parameters field-name ambiguity, but it communicates enough for an agent to invoke the tool and interpret the core behavior.

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?

The schema already documents all parameters, so the baseline is 3, but the description adds meaningful context by explaining compute mode behavior, gpu:true delegation rules, chain provenance via execution_hash, and the fact that policy_parameters may be computed server-side. The only notable gap is that actual policy_parameters field names are deferred to a manifest, not stated inline.

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 name and first line identify a Basel 3.1 reporting delta calculator and an OpenChainGraph compute node in the capital_assessment cluster, so an agent can distinguish it from unrelated compute tools. It does not, however, state with a precise verb what 'delta' is computed between (e.g., before/after Basel 3.1, actual vs expected RWA), which leaves some ambiguity.

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 explains operational mechanics such as compute modes and synthetic-input requirements, but provides no explicit guidance on when to choose this tool over siblings like compare_basel_2023_vs_2026 or compute_rwa_erba_2026. There is no 'use this when' or 'prefer X instead' statement, so the agent must infer the tool's applicability from its name and output feeds.

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