signin_failure_stats

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

With no annotations provided, the description bears full responsibility and delivers richly: it declares read-only status with specific permission/role requirements (AuditLog.Read.All or Reports Reader), discloses that Graph cannot filter server-side so it walks up to max_pages and aggregates client-side, and explains that capped=true means counts are a sample rather than a census. It also states the hours clamp (1–720). This exceeds what annotations would typically convey.

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 text is front-loaded with the core purpose and output views, then layers behavioral detail and Args. It is dense but not bloated; the sentences on KeyCloak mirroring and the MCP Server/Graph gap, while adding institutional context, are slightly extraneous to actually invoking the tool. Overall well-organized with a consistent summary → behavior → args structure.

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?

For a tool with 0% schema coverage, no output schema, and no annotations, the description is highly complete: it names all four output views, the spray_suspects flag with its threshold, the capped sampling caveat, auth requirements, and the pagination semantics. The only minor omission is a precise shape of the returned views (exact fields per row), but the narrative gives enough for an agent to call it and interpret results.

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%, so the description must carry parameter docs, and it does: hours ('How far back to look, clamped to [1, 720] (30 days)') and max_pages ('Page budget (default: ENTRAADM_MAX_PAGES_DEFAULT)'). It also connects max_pages to the behavioral capped flag, adding causal meaning absent from the bare schema. The minor schema/description default mismatch (null vs ENTRAADM_MAX_PAGES_DEFAULT) is benign since null triggers the env-var default.

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 names a specific verb ('aggregates') and resource ('failed sign-ins across the whole tenant') and enumerates exactly four output views (AADSTS error codes, failing users, applications, source IPs). It also positions itself against siblings as the 'Entra ID counterpart to the RADIUS failure patrol' and explicitly references signin_logs for error-code meaning annotations, so an agent can distinguish it from neighboring tools without opening schemas.

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 gives a clear driving use case: detecting low-and-slow password sprays that evade per-account smart lockout, capping with the concrete 5-distinct-users spray_suspects threshold, and noting that neither the official Microsoft MCP Server nor Graph offers this aggregation. It does not, however, explicitly state when NOT to use it or name the alternative tool (e.g., use signin_logs for raw per-sign-in detail), so a small exclusion gap remains.

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