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

Annotations already declare a non-read, non-idempotent, non-destructive write; the description adds what annotations cannot: no money moves and no guardrail changes, entries are visible to other sessions/clients, and empty values clear settings/schedule/notes (a real destructive side effect the annotations do not capture). The per-user cap of 20 active/paused is disclosed up front rather than as a runtime error.

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?

One orienting sentence followed by mode-labelled bullets, each carrying a distinct parameter recipe. No restatement of the name or title, no filler; the constraint and safety statements are placed at the end where they close the loop.

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 an 8-parameter multi-mode upsert tool, everything an agent needs before calling is present: mode selection, required combinations, id sources, clearing semantics, quota, and side-effect scope. Return values need not be described because an output schema exists.

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 would be 3, but the description adds mode-conditional semantics the schema cannot express: which parameter combination is valid for each operation, that name defaults from the starter card, what belongs in settings (amounts, mix, exits, ask-first flag), that schedule is plain words, and that last_run_at is ISO 8601. This is meaning beyond the flat field list.

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+resource ("Record what you run") plus the reason (other sessions and clients see it), and immediately decomposes the tool into its five operating modes (START / PAUSE-RESUME-STOP / EDIT / HEARTBEAT). Sibling read tools (list_strategies, get_my_active_strategies) are named as the source of ids, so an agent can tell this is the write-side counterpart.

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 explicit when-to-use rules per mode: omit active_strategy_id to start, pass it to pause/resume/stop/edit/heartbeat, and pass only changed fields on edit. It also names the alternative sources of ids and the capacity constraint ("At most 20 active or paused per user; stop one to start another"), which is the exclusion an agent actually needs.

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