inventory_alerts

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

The description is exceptionally transparent: 'Pure computation; no persistent alert store in Phase 1. Empty is an ok empty list. Posts nothing.' This discloses the read-only, stateless nature beyond the readOnlyHint annotation. It also notes the optional future acknowledgement feature (stable alert_key), setting expectations about current behavior. This fully aligns with the annotation, adding context about persistence and side effects.

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 description is front-loaded with the core purpose and then details the output structure, filters, and behavioral notes. Each sentence adds value, but it is somewhat long and could be tightened. For example, the list of alert types could be shortened, and the parenthetical spec reference '(US-J07.9)' is non-essential. Overall, it is well-organized but slightly verbose.

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 the absence of an output schema and parameter descriptions, the description carries a heavy burden. It covers the return structure (alert fields), filter options, and behavioral guarantees (read-only, no persistence, empty list). It does not explain the 'thresholds' parameter, which is a notable omission for a tool that aggregates alerts. It also doesn't provide examples or enumerate valid severity/type values, but these might be inferred from context. Overall, it is comprehensive but not fully complete.

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?

The schema has 0% coverage for parameter descriptions, so the description must compensate. It mentions 'Optional severity / types / limit filters', which covers three of the five parameters (limit, types, severity), but omits 'thresholds' and the required 'workspaceId'. While it implies these filters exist, it does not explain their values or formats. The description adds some meaning but leaves gaps, especially for thresholds, which could be crucial for tuning alerts.

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 clearly states the tool's purpose: a unified alert feed that merges multiple alert types into a single, de-duplicated, severity-prioritized list. It names the specific alert categories (low-stock, anomalies, overdue stocktakes, valuation drift, near-expiry lots) and lists the output fields. This distinguishes it from sibling tools like detect_anomalies or inventory_anomalies, which are more specialized. The verb 'poll' and resource 'everything that currently needs attention' make it unambiguous.

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 description explicitly says it is 'one tool an agent can poll for everything that currently needs attention', which is a clear when-to-use statement. It also mentions optional filters (severity, types, limit) to refine results. However, it does not explicitly name alternative tools or conditions when NOT to use it, such as when a specific anomaly detection is needed instead. The guidance is implicit rather than comparative.

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