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

Annotations already cover readOnly/idempotent/non-destructive, and the description goes meaningfully beyond them: inputs are processed transiently and never stored or logged, compute:'auto' resolves server-side for gpu:false nodes with a registered kernel while 'browser' returns a delegation URL, gpu:true always delegates, and an AP2 artifact with execution_hash is exported for provenance. That is real behavioral context an agent cannot infer from annotations. Consistent with readOnlyHint=true (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?

Badly mis-front-loaded: the actual purpose appears only in the trailing 'Summary' paragraph, after boilerplate about OpenChainGraph compute nodes, transparent processing, AP2 provenance, a full documentation URL, and an FV-status URL with a 64-character hash. Several sentences (the snapshot-receipt URL, 'this receipt verifies offline regardless of whether that file is ever fetched') do nothing to help tool selection or invocation.

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 compute node with no output schema, the description does supply the return surface (basis_risk_delta_nii, is_material, material_threshold_pct, index_results, etc.), the compute/provenance behavior, and the beta semantics of index_exposures. An agent has enough to call it correctly, though the truncation ('indices not...') and the diffuse layout leave minor gaps.

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 baseline is 3. The description does add one genuinely useful semantic the schema omits: index_exposures are priced indices each carrying a caller-declared historical beta versus the reference rate, which clarifies the untyped array items. Beyond that it mostly restates the compute enum already documented in 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 Summary sentence is precise: a Comptroller's Handbook IRR basis-risk delta-NII calculator that sweeps one reference-rate shock across multiple priced indices with caller-declared betas, isolating incremental delta-NII. That is a specific verb+resource+method. It does not, however, distinguish itself from close siblings such as evaluate_irrbb_sot_nii, calculate_irrbb_eve_shocks, or run_rate_shock_ladder, so an agent cannot route on purpose alone.

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 only usage-adjacent statements are the privacy constraint ('Use synthetic or anonymised inputs only') and the compute-mode default explanation. Neither tells the agent when to pick this tool versus the standardised-approach NII or EVE siblings, and no preconditions or exclusions are given.

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