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

Annotations already cover the read-only/idempotent/non-destructive profile, and the description goes well beyond them by disclosing the default filter behavior, the meaning and trust level of `suggested` vs `confirmed` rows, the conflict flag, the 30-day spend snapshot, the explicit exclusion of code/file paths/excerpts, and the caveat that `confidence` is a score not a probability. It also names the underlying endpoint (GET /api/accounts/:accountId/repo-mappings) for traceability.

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 key facts are front-loaded in the first sentence (what is listed and what fields come back), with trust rules and exclusions following. It is dense and slightly overloaded with parenthetical enumerations, but nearly every clause supplies information an agent would otherwise have to guess at.

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 usefully inventories the returned fields and explicitly says what is not returned, and it clarifies status and confidence semantics. The remaining gap is the four undocumented filter/pagination parameters, which matters less here because all parameters are optional and the tool works with zero arguments.

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 0% for six parameters. The description adds real meaning for two of them — the `status` default and the suggested/confirmed distinction, and the 0-1 nature of confidence relevant to `min_confidence` — but `limit`, `offset`, `repository_id` and `cost_source_id` receive no explanation at all, leaving the agent to infer their filtering and pagination semantics.

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 opening clause names a specific verb and resource pair — listing which code repository (or monorepo service directory) owns which cost entity — and enumerates the entity kinds (resource, Kubernetes workload/namespace, tag value, service, linked account) plus the fields returned. No sibling tool in the list does repo-to-cost mapping, and the description makes that distinct role obvious without requiring a schema lookup.

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 concrete usage context: the `status` filter defaults to `confirmed`, `suggested` rows are undecided machine proposals that 'must not be quoted as facts', and confirming/rejecting are app-only so this tool cannot mutate state. That is strong guidance on how to interpret results, though it never explicitly frames when to reach for this tool versus querying cost data directly.

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