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

Annotations already declare destructiveHint=true, idempotentHint=false and openWorldHint=true, yet the description adds substantial context: the save/backup/close/reopen side effect with a ~10-20 s duration and an instruction to warn the user, the all-or-nothing atomic contract, the enumerated rejection codes with extra payload (did_you_mean, valid ids), and the absence of per-batch undo. The retry/dedup explanation is also consistent with idempotentHint=false (dedup only with an explicit batch_key) rather than contradicting it. Slightly short of 5 only because return shape and what a successful response looks like is left implicit.

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?

Dense but well front-loaded: capability, then input sources, then live-edit side effect, then CONTRACT/RETRY/ordering sections in labeled blocks. The op-type parenthetical list is long, and the trailing '…' is a minor untidiness, but nearly every sentence carries operational information an agent would otherwise have to guess.

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 destructive, open-world batch-mutation tool with no output schema, the description covers the failure contract, error codes, atomicity, retry/dedup, ordering, absence of undo, and a verification path. Combined with a partially described schema, an agent has everything it needs to call this safely or route the agent to the right reference tool.

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 75%, so the schema already documents ops, atomic and batch_key; the description still adds real meaning by spelling out the dedup-on-retry semantics of batch_key and the 'stores the good ops, rest rejected' behavior of atomic:false. It does not document session_id, but that is self-evident from the naming and the pointer to doseedo_get_session.

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?

Opens with a specific verb+resource ('Edit a DAW session') and immediately enumerates the edit surface (regions, MIDI notes, sampler samples, tracks, mixer, plugins, tempo, markers, key). It is clearly distinguishable from the read-side siblings doseedo_get_session and the documentation-only doseedo_edit_ops_reference, which it names explicitly.

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?

States precisely where inputs come from ('ids and current values from doseedo_get_session, op arguments from doseedo_edit_ops_reference') and where to verify results ('verify with doseedo_download_session'). It also gives the conditional branch for sync.live_edits true vs false, so the agent knows when the call mutates the live Logic project versus a stored copy.

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