toolbox

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

No annotations are provided, so the description carries the full burden. It discloses that output is paged and preserved, that describe can be compact, and that profile management exists. However, it does not reveal side effects of enable/disable/profile operations (like whether they alter persistent state), auth requirements, or any output format details. For a tool that can modify bindings and profiles, more transparency is needed, but the basic behavioral facts are present.

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 dense but poorly structured: it front-loads operational guidance but then lists operations and profiles in an unorganized way. Each sentence adds info, but the flow is jumbled, mixing instructions, operation list, and profile list without clear separation. It's not concise enough for the complexity; ideally it would use a structured list for operations and profiles.

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 the tool's complexity (12 parameters, multiple operations, nested objects), the description covers the high-level workflow and mentions key parameters and profiles. However, it lacks details on return values (no output schema), side effects of enable/disable/profile, and edge cases like how output_id relates to output operation. It is minimally adequate but leaves gaps that an agent might need to infer or test.

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 100%, so each parameter already has a description in the schema. The description adds some usage context around certain parameters (e.g., 'compact' and 'tool_operation' relationship, 'profile' list), but it doesn't add meaning beyond the schema for most parameters. It does clarify the default operation is 'list' and that 'schema_handle' might return a short acknowledgement, but that's marginal. With full coverage, baseline 3 is appropriate.

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 lists multiple operations ('list', 'describe', 'call', etc.) and the tool name 'toolbox' hints at a dispatch tool, but the core purpose is not stated as a single verb+resource. It bundles several operations, making the primary purpose ambiguous. The list of sibling tools (read, write, edit) are specific file operations, while this one is a general dispatcher, but that distinction is implied rather than explicit.

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 gives some guidance on when to use certain operations: 'Use describe, preferably compact with tool_operation, then call with arguments' and 'output retrieves preserved paged command output'. However, it does not explicitly contrast with alternatives or state exclusions. It implies that for performance, compact describe is preferred, but it doesn't say when NOT to use this tool vs. siblings; it assumes the agent knows this is a dispatch tool.

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