analyze_envelope

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

With no annotations, the description carries the full burden and excels. It discloses the intentional U9 fix (mean subtraction/window) and why it exists ('DC leakage used to bury the low-frequency FTF zone'), states invalid bands raise ValueError and are 'NEVER silently clamped', notes the band is echoed in the result, and explains the deterministic default (leading 1.0-s segment) and random_seed behavior. Also clarifies no reference frequencies are assumed, preventing false expectations.

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 description is long but every sentence earns its place: algorithm, rationale, requirements, constraints, defaults, and comparison guidance. It uses a clear structure (intro, behavior, Args, Returns, Raises) and front-loads the purpose. No fluff or repetition; the 'THE unified envelope tool' line, while emphatic, reinforces its role among siblings.

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?

The tool is complex (6 parameters, prerequisites, error conditions, segment selection) and the description covers all aspects: preconditions (signal must be loaded), constraints (band vs Nyquist), deterministic vs random behavior, no assumed bearing frequencies, and error types. The output schema exists, so it appropriately keeps return details brief ('EnvelopeResult with the band actually used, top peaks, and comparison guidance').

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%, so the description must fully compensate. It explains every parameter: signal_id (ID from load_signal), filter_low/filter_high (edges in Hz with defaults, Nyquist constraint), num_peaks (count of top peaks), segment_duration (duration, default leading 1.0s, None for full signal), random_seed (seed for random position, None = deterministic). It also adds error semantics for invalid values, going far beyond the bare schema.

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 description states a specific verb+resource: 'Envelope-spectrum analysis of a stored signal (bearing fault screening)' and details the algorithm ('bandpass filter -> Hilbert envelope -> mean subtraction + Hann window -> FFT -> top peaks'). It clearly distinguishes itself from siblings like analyze_fft and check_bearing_faults by positioning itself as 'THE unified envelope tool' and referencing the comparison tools for subsequent analysis.

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?

Provides explicit usage context: 'Requires the signal loaded via load_signal() first', explains when to use the tool (bearing fault screening), and gives alternative/next-step guidance: 'compare the returned peaks against frequencies computed for the actual bearing and shaft speed (check_bearing_faults or calculate_bearing_characteristic_frequencies)'. It also clarifies segment selection options and when to pass None or random_seed.

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