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

Annotations already declare non-read-only, non-destructive, non-idempotent, closed-world, so the bar is lower; the description still adds meaningful behavior: writes require expectedRevision (optimistic concurrency), and changes touching Recording, Safety or a reviewed release block on a human in Tasks. It omits what rollback/revisions actually return and whether edits are reversible, which would have completed the picture.

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 and the agentMemory routing are correctly front-loaded, and the operation map is dense with no filler. However, the telegraphic sentence fragments ('operation list.', 'revisions: the history.') and the quoted-empty-string convention for defaults cost more in parseability than they save in 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 multiplexed tool with 8 operations, 18 parameters, no output schema and near-zero schema documentation, the description covers the operation/parameter mapping and the approval workflow well. It still leaves several parameters undocumented and never indicates what list/get/revisions return, which an agent needs in order to sequence calls.

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 only 11%, so the description must compensate and largely does: it maps each operation value to its parameters (get→skill/section/newest, errors→tool/errorCode/topic, update→mode/body/to, edit→body/preferences/mode trim, rollback→to, progress→step/state/note, plus expectedRevision). It leaves limit, summary and companyId entirely unexplained, which is a real gap across 18 parameters.

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 resource ('this company's own playbook for each job') and gives an operative verb ('Read the skill before the work and follow it'), which lets an agent distinguish it from sibling agentMemory by explicitly assigning facts there instead. The 'DeepLedger' branding and the fragmented phrasing keep it from being instantly parseable, but the purpose is unambiguous.

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?

It states when to use it ('read the skill before the work') and routes the competing concern to a sibling ('facts stay in agentMemory'), which is real when/when-not guidance. It does not spell out when to skip the skill lookup or which operations are read-only versus mutating, so it falls short of explicit exclusions.

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