dashboard_overview

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

Goes far beyond the readOnlyHint annotation: declares it 'never cached (pure read, F00 posts nothing and owns no tables),' details per-tile permission gating (omitted[] with per-tile gates such as read_books and time.read), and discloses degradation behavior (ok:false with codes like needs_vat_config), empty-workspace behavior, and from>to invalid_range semantics. Also discloses that savedViewId resolution 'can never widen access.' There is no contradiction with annotations — the description reinforces readOnlyHint.

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?

Front-loaded with the core definition and scoping constraints, then enumerates tiles and behavioral edge cases roughly in order of importance. Every clause carries routing, gating, or error semantics; the tile-to-source enumeration is long but earns its place. The single dense paragraph is harder to scan at a glance than a structured list, which costs one point.

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 no output schema and zero schema-level param descriptions, the description is remarkably complete: tile composition, units (integer-Rappen/basis points), source mapping, permission gating, degradation codes, empty state, invalid_range, and saved-view fallback are all covered in prose. Genuine gaps remain: the exact date string format for from/to and the top-level response envelope are not specified.

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 carries the burden, and it compensates well: from/to get date-range semantics plus the ordering constraint (from > to answers invalid_range), and savedViewId gets rich semantics (G00 saved view, entityKind 'workspace', layout 'dashboard', restriction/ordering, viewFallback:true on failure). workspaceId is only implicit (as the scope of A24 role gating and empty-workspace behavior), leaving a small gap.

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 by identifying the resource precisely — 'one KPI tile wall for a date range' — and names its eight constituent tiles with their exact data sources (income_statement, aging_report, vat_return, stock_valuation_report, etc.). This distinguishes it from sibling source-report tools and from per-tile tools like dashboard_tile without opening any schema.

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?

Explicitly routes drill-down behavior to the source tools: each tile 'embeds a drill descriptor (studioRoute, mcpTool, params) naming the EXISTING source tool behind the number: call that tool for the rows.' It also specifies when saved views apply and the fallback when a view cannot be resolved. It lacks an explicit when-not-to-use statement against the closest sibling (dashboard_tile), so exclusions are absent.

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