batch_update_opportunity

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

Annotations only declare readOnlyHint=false and destructiveHint=false, leaving the description to carry the behavioral burden — and it delivers. It discloses concurrency behavior with a configurable cap ('default cap 5 (CAPSULE_MCP_BATCH_CONCURRENCY)'), the exact return shape, and critically the failure semantics: 'Partial failures possible; Capsule has no rollback.' For a batch mutation, this is precisely the risk information an agent needs.

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?

Five sentences, zero fluff, and the core action is front-loaded. Each subsequent sentence adds a distinct piece: shape relation, use case, concurrency behavior, return contract, failure mode. It earns a 4 rather than a 5 because the single dense paragraph could be slightly more scannable — the concurrency and failure-semantics details could be visually separated — though nothing approachable is wasted.

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?

Given there is no output schema, the description correctly documents the return contract ({ results: [...], summary: {...} }), the input-shape relationship to the single-update sibling, the 1-50 cardinality, and idempotency-adjacent safety info (no rollback). This is complete for a batch utility whose error contract matches its single sibling. Minor deduction: the contents of the per-item `ok` object are left undefined, and there's no note about whether a single malformed item aborts the batch or gets reported inline.

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% and the items parameter already carries a rich description explaining the 1-50 bound and the rationale ('so a single tool call can't burn an outsized share of Capsule's hourly per-token rate budget'). The main description complements this by documenting the return shape. Slight deduction because the 'same shape as update_opportunity' reference assumes the agent cross-references another tool's schema, and field-level semantics are delegated entirely.

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?

Opens with 'Update 1–50 opportunities in parallel' — a specific verb, resource, and scope qualifier in one breath. It then names the sibling it is not ('Same input shape as update_opportunity but wrapped in an items array'), making the single-vs-batch distinction unmistakable against siblings like update_opportunity, batch_update_party, and batch_update_project.

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?

'Use this — not N sequential update_opportunity calls — for mass stage transitions (e.g. move a milestone batch to Won), owner reassignments, or value adjustments' is an explicit when-to-use directive that names the alternative and gives concrete example scenarios. This is the gold standard the rubric calls for.

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