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

Annotations only supply low-information mutation hints (readOnly=false, idempotent=false), so the description carries real weight: 15-minute URL expiry, 1-hour default handle TTL, single-use handle semantics, and two concrete error codes (UNREADABLE_FILE_REFERENCE, UPLOAD_TOKEN_EXPIRED) with the re-issue behavior. This is exactly the disclosure the annotations cannot provide.

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 purpose and return values before the mechanism details; nothing is filler. The second and third sentences are densely stacked with parentheticals, making them slightly run-on, but the size is justified by the tool's four-path complexity.

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?

Output schema exists so return formats need not be restated, yet the description still calls out the key returned fields. Combined with error codes, TTL defaults, and per-path constraints, an agent has everything needed to stage rows correctly and hand off to ingest_accounts.

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 compensate, and it largely does: csv_file (host file, local path fails), csv_text (raw CSV, 16 MB cap, newlines preserved), rows (chunks appended via upload_handle), upload_handle (single-use). ttl_seconds is only implied by the 'expires after 1 hour by default' clause rather than named, leaving one parameter short of fully covered.

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?

States a specific verb and resource ('Stages account rows for ingest_accounts') plus the exact artifacts returned (upload_handle, csv_headers, suggested_ingest args, geocode_plan). It is cleanly distinguished from its consumer sibling ingest_accounts, which claims the handle.

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?

Names the consuming tool and lays out four ingestion paths with selection conditions ('POST ... fastest for large files'; csv_text 'up to 16 MB'). Missing an explicit when-not-to-use rule, but the workflow placement relative to ingest_accounts is unambiguous.

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