run_boundary_attack

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

With no annotations provided, the description carries the full burden. It discloses the iterative nature, the condition for stopping (prediction flip or max_steps), the batch cap limit (100), and the sampling from same batch. However, it does not specify permissions, side effects, or return format, but for an adversarial attack tool, the disclosed behavior is adequate.

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 consists of two non-redundant sentences that front-load the key purpose and behavior. It is concise with no filler, though some minor detail on parameter roles could be added without breaking conciseness.

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?

Given four parameters, no output schema, and no annotations, the description explains the algorithm and constraints but lacks complete parameter descriptions and return value information. It covers the core operation but leaves gaps in parameter semantics, making it moderately complete.

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 description coverage is 0%, so the description must add parameter meaning. It only explains the 'batch' parameter indirectly (capped at 100) and mentions 'max_steps' but not 'step_size' or 'model_handle'. This leaves two parameters undocumented, providing insufficient semantic context beyond the 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 clearly states the tool runs an iterative decision-boundary attack, describing the process of moving samples toward opposite-predicted-class samples until prediction flips or max_steps exhausted. It also mentions the batch cap, distinguishing it from sibling tools like run_fgsm which use different attack methods.

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?

The description explains the tool's operation and limits but does not explicitly state when to use this attack versus alternatives (e.g., run_fgsm) or when not to use it. Guidelines for context are implied by the algorithm description but lack direct comparison.

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