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

Annotations already cover read-only, idempotent, non-destructive, closed-world semantics, and the description adds genuinely new behavior: compute:'auto' vs 'browser' delegation, GPU nodes always delegating, transient non-retained processing, and an AP2 artifact with execution_hash for chain provenance. That is meaningful operational context beyond the structured annotations, though the compute-mode text largely duplicates the schema's own parameter description.

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 statement is buried behind regulatory deadline, compute-node boilerplate, a documentation URL, and a lengthy FV-status/receipt paragraph that mostly advertises offline verification. Several passages (compute mode, output enumeration) restate schema or structured data. Front-loading the 'Summary: Replays...' sentence would materially improve selection.

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 does the necessary work of enumerating return fields (trough_capital_mn, trough_capital_ratio_pct, scenario_set_digest, not_a_submission, etc.) and flagging the artifact/provenance output. For a nested policy_parameters tool it is nearly complete; only the parameter-level detail (units, valid scenario values) is thin.

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 compute, parent_hashes, parent_tool_ids, and policy_parameters semantics are already documented in the schema. The description adds only a high-level mention of loss and PPNR coefficient functions and required policy keys, without explaining units, formats, or how the scenario selector interacts with the Fed paths. Baseline 3 is appropriate.

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 gives a specific verb, resource, and scope: replaying the Fed's published 2026 28-variable supervisory paths against user-supplied loss/PPNR functions to produce a quarterly P&L and capital walk. That is far more precise than the name alone. However, it never contrasts itself with near siblings like compute_stress_test_scenarios or calculate_irrbb_eve_shocks, so differentiation is left to inference.

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 only implied through context (capital_assessment node, 2027-02-01 annual re-pin deadline, 'Use synthetic or anonymised inputs only'). There is no statement of when to choose this tool versus the alternative stress/scenario tools in the sibling set. The 'Output feeds' line describes downstream lanes, not selection criteria.

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