Keep a note the person said yes to

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

Annotations only declare the write/idempotent/safe profile; the description adds substantial behavior beyond them: the resulting Memory record is tagged 'unverified' consent, the Sentinel-held case is excluded, 'all' expands to every pending note as separate ledger lines, and a hard cap of 50 keeps per local day resetting at the vault machine's midnight with two documented ways around it.

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 gating condition is front-loaded in the first clause and every sentence carries information (consent rule, result state, exclusion, all-branch, cap, workarounds). It is dense and slightly run-on in places ('A note Sentinel is holding is not kept this way'), which costs a point but no filler is present.

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 and a single documented parameter, the description covers the consent gate, the resulting side effects, the batch case, the rate limit, and the escape hatches — enough to invoke correctly. It stops short of saying what the call returns or what a failed cap does to the note's pending state.

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% and already documents id / last-8 / "all", so the baseline applies. The description still adds meaning the schema lacks: what the "all" branch actually does ('keeps every note pending lists, when the person said keep all; each is its own ledger line') and that cap counting applies per note.

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 — keep a staged note that the person consented to — and immediately names the resulting state ('The note becomes Memory'). It also carves out cases handled elsewhere ('A note Sentinel is holding is not kept this way'), so an agent can distinguish it from stage/drop/handoff without opening schemas.

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?

Gives an explicit precondition ('only when the person has said yes to this note in their own words in this conversation'), an explicit prohibition ('never keep on your own judgment', 'never draw a panel'), and names the alternative escalation paths ('magnemo yes <id>' or the client's one-question handoff). When-to-use, when-not-to-use, and alternatives are all present.

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