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

Annotations only declare the generic safety profile (non-read-only, open-world, non-idempotent, non-destructive). The description adds the genuinely useful traits: GPU-backed vs CPU detector selection, the automatic fallback, the ~10-minute input limit, synchronous ~5–30 s execution, and an explicit 'Cancel: not applicable — result is returned in the submit response.' That last point is critical because doseedo_cancel_job exists as a sibling and an agent would otherwise try to cancel.

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-loaded with the payload shape, then detector defaults, then input sourcing, then limits and timing. Dense parentheticals make it slightly run-on, but virtually every clause carries operational information (fallback, 10-min cap, sync timing, no-cancel).

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?

There is no output schema, so the description correctly enumerates the return fields (chord_spans with confidence/beats, chords beat map, bpm, beats_per_bar, beat_map). Combined with the sync/cancel clarification and input limits, an agent has everything needed to call this correctly in one shot.

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 baseline is 3; the description still earns credit by explaining the default value semantics of deep (deep ACE with automatic CPU fallback) and the distinction between audio_key (an upload, ingest/) and audio_url (a prior result URL as returned, or any public https URL). These are behavioral implications the schema descriptions only partially convey.

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 (chord detection on audio) and immediately disambiguates the output from siblings like doseedo_beat_grid and doseedo_tempo_map by naming chord_spans, chords, bpm, beats_per_bar and beat_map. An agent can tell exactly which analysis product this returns without opening the schema.

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 concrete usage conditions: default is the deep ACE detector with CPU chroma fallback, deep:false forces the fast detector, supply audio_key or audio_url, 10-minute cap, and 'works best on a harmonic stem (piano/guitar/bass) or a clean mix' — which implicitly points at doseedo_separate_stems. It never explicitly names a sibling it replaces or excludes versus doseedo_analyze_song/doseedo_beat_grid, so it stops short of full routing guidance.

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