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

Annotations only provide readOnly=false / destructive=true / openWorld=false; the description adds substantially more: the call is two-phase (stages a request, owner reviews and approves via browser link with an agent PIN), 'none' means hidden and not served, and there is an explicit security rule never to accept the PIN in chat. That is exactly the behavioral context annotations can't express.

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-loads the actual semantics (the three levels) before the workflow and safety notes, and every subsequent sentence carries operational weight. It is dense but not padded; the colon list in sentence one is slightly compressed but readable.

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?

With no output schema, the description must tell the agent what the call actually does — and it does, making clear the change is staged and only takes effect after owner approval, plus who can do what and which tool supplies valid skill identifiers. Nothing needed to call or interpret it correctly is missing.

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 100%, so all four parameters and the level enum are already documented in the schema; the description's restatement of the three levels and the 'list_skills shows the slugs' pointer largely duplicate it. Baseline 3 is appropriate, with only marginal added meaning (e.g. 'not served' sharpening 'none').

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 opening sentence gives a specific verb+resource at the exact granularity of the tool (one access role × one skill) and enumerates the three possible states. That granularity distinguishes it from the nearby set_access_role_capability / set_access_role_mailbox_access / assign_access_role siblings without either schema being opened.

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 explains the precondition (skill slugs come from list_skills), the default state (use for every role), the owners-always-edit exception, and the post-call workflow the agent must expect. It does not explicitly name which sibling to reach for instead when the intent is a non-skill capability, so it stops short of a 5.

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