logic_level

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

With no annotations, the description carries the full behavioral burden and does so well: it discloses that 'plan' applies nothing, that 'apply' mutates track faders and verifies in dB, that dry_run suppresses the write, and — critically — that relative mode is NOT idempotent and double-applies on a second run. That non-idempotency warning is exactly the kind of trait annotations could not convey. It stops short of saying whether fader moves are reversible or what the measurement output looks like.

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?

Well front-loaded: purpose first, then the action table, then the mode caveat that governs correctness. The action list is compact and scannable. The 'which is what a human engineer does anyway' aside is minor filler, but the rest earns its space.

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 10-parameter, four-action, no-annotation, no-output-schema tool, the description covers the behavioral decision surface well but leaves real gaps: six parameters undescribed and no indication of what analyse actually returns to the agent. An agent can choose the right action, but not confidently populate the tuning parameters or interpret the result.

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 0% across 10 parameters, so the description must compensate, and it only covers roughly half: action (all four values), mode (both values plus the default's implications), dry_run, and paths. apply_pan, directory, reference, headroom_db, max_move_db and max_seconds are never mentioned, leaving six numeric/path parameters completely undocumented in both schema and prose.

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 states a specific verb and resource: measuring rendered audio and balancing the mix. The four action values are enumerated with a one-line explanation each, so the tool's scope is unambiguous. It never names a sibling (logic_export, logic_offline or logic_tracks could plausibly overlap), so differentiation must be inferred from the action list rather than being stated.

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?

Explicitly states when to pick each action: 'analyse' for measurement only, 'plan' when nothing should be applied, 'export_and_plan' when stems must first come out of Logic, 'apply' to commit. The mode section goes further, naming the exact condition that selects relative vs absolute ('right when the stems were rendered through those faders' vs 'stems that came from outside this project'). This is when-to-use guidance at the level of a decision tree.

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