document

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

No annotations exist, so the description carries full burden. It DOES disclose the most critical behavioral subtlety — that preflight never accepts file_paths — which is genuinely valuable. However, for an op named 'auto' that ingests or checks readiness, and operations like 'prepare_ai' or 'export_assets', it never reveals side effects (file writes, DB mutation, storage consumption), return shapes, or state changes. The description's only behavioral claims are about parameter rejection, not about what happens to the document at runtime.

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 bullet-based routing table is information-dense with no filler — every line adds a constraint or maps an operation. However, the description opens with a code block of routing contracts before telling the reader what any of these operations actually do, which makes it dense but not front-loaded. The final paragraph is genuinely useful context but buries the lede about sibling coexistence. It's terse for sure, but terseness without semantic anchors risks confusing an agent that doesn't already know the domain.

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 facade with 8+ operations, 0% annotation coverage, 0 output schema, and a 20% schema description rate, the description only partially compensates: it nails the contract routing and preflight edge case but leaves out return values, error conditions, idempotency, and whether operations like export_assets or agent_assets write to disk or return bytes. An agent deciding whether to call inspect versus fetch_document_asset gets no guidance here on outcome, only on parameters.

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?

With only 20% schema coverage, the description carries heavy weight — and it does add the crucial op-dependent contract layer that no schema could express (doc_id is required for inspect/prepare_ai but rejected for preflight). That cross-parameter conditional logic is the single most important thing an agent needs to know, and the description provides it. Nonetheless, the remaining 80% of parameters (criteria, deskew, index_knowledge_graph, marker_max_pages_per_chunk, etc.) go completely unexplained in both the description and schema.

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 nails a core purpose marker: 'Consolidated PDF document entrypoint'. The verb 'Consolidated... entrypoint' combined with the explicit op→params contract table makes the facade intent unmistakable, and the line 'Existing direct document tools stay registered' clearly positions it as an alternative to the sibling direct-tools. It loses a point because it never defines what each operation (auto, preflight, inspect, prepare_ai, etc.) actually accomplishes functionally — but as a router for PDF operations, the purpose is quite legible. The textual contract maps cleanly onto the sibling tool names (ingest, convert_document, etc.) even without explaining semantics.

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 unusually specific routing rules: 'preflight: pdf_path only; file_paths is rejected' and 'auto: exactly one of file_paths (ingest) or doc_id (readiness)'. It even gives a migration note in the last paragraph. But it stops short of saying 'use this facade when...' or 'prefer this over ingest_documents if...' — the line 'Existing direct document tools stay registered' implies the choice but never gives the agent a decision rule for when to use the facade versus a direct sibling.

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