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

Annotations already cover read-only/idempotent/closed-world, but the description adds substantive traits they cannot: it copies existing org evaluation claims and never invents CE or claim values, org narrative fields are untrusted per ADR 0008, Wave BI provenance is attached to responses, and 'no figure, rating or order is added.' That is a real trust and provenance model an agent needs when quoting figures.

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?

Purpose and index location are front-loaded, and the safety constraints come before the response-shape detail. There is mild redundancy between 'never invents CE / claim values' and 'No figure, rating or order is added,' and the single dense paragraph would scan better as separate sentences.

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 has to carry return semantics, and it does describe the decision_brief payload (donate_url next_action, evaluator's fund alternative, like-for-like comparison page, modeled/guaranteed flags) plus provenance. It stops short of describing the claim record fields an agent would receive for each row.

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 the baseline would be 3, but the description adds meaning beyond the schema for `verified` — use verified:true with a non-null value when quoting cost-effectiveness — which the schema's plain 'filter on claim.verified' does not convey.

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 states a specific verb and resource ('Read-only flat claims index (data/claims_index.json)') plus the optional filters, which is far more concrete than the bare name/title. It never names the sibling get_claim for single-claim retrieval, so an agent must infer the list-vs-single distinction from the name 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?

It gives actionable guidance for filter choice: 'Prefer verified:true + non-null value when quoting cost-effectiveness.' That is a clear usage rule, but there is no statement of when NOT to use this tool or which sibling (get_claim, list_charities, search_orgs) to prefer for adjacent needs.

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