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

Annotations already flag destructive=true, idempotent=false, but the description goes well beyond: it discloses optimistic concurrency (refused when the record changed since baseVersion), refusal on unknown paths or schema violations, the atomicity guarantee ("a refused save stores nothing"), that unknown fields are silently dropped, null removes a target, and that out-of-range indices append. These are precisely the destructive and non-idempotent behaviours an agent must anticipate.

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 purpose is front-loaded, but the body is a dense run of semicolon-chained clauses ("...indexes from one past a list's last item add items there, in turn, a null value removes what the path names, and no path lies inside another") that forces re-reading. Nearly every clause carries information, but paragraph-level structure is poor for a description of this length.

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 an enormous nested record schema and no output schema, the description does the heavy lifting: it defines the change model, the refusal model, and describes the return shape ({version, stored, changed, hints}) in enough detail to be usable. The one gap is routing relative to the other write/import/restore siblings.

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 coverage is 100%, so the baseline is 3, and the description adds real meaning the schema does not: path grammar (header, a whole section, comments[3], roles[0].projects[2]), append-on-out-of-range-index, null-for-delete, and the no-overlapping-paths constraint. It also explains baseVersion's role and what the note is for, though the note's purpose is largely restated from the 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?

States a specific verb and resource: "Store changes to the career record", and specifies that changes are path/value pairs against a baseVersion. An agent knows exactly what operation this is. However, it never names the near neighbours (import_career_record, restore_career_version) that also write the record, so the distinction is by inference rather than stated.

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?

Usage is implied by the mechanics (supply baseVersion from get_career_record, paths from the returned record), which effectively tells an agent when this tool is applicable. But there is no explicit when-to-use versus import_career_record or restore_career_version, and no statement of prerequisites such as needing a fresh read first, which is exactly the situation where an agent would pick the wrong sibling.

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