dsh_cancel

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

Beyond the destructive and non-read-only annotations, the description reveals critical behavioral nuances: the non-atomic removal for queue scope, the reliance on a 'latest snapshot', and the preservation of queued work when canceling the active turn. These details significantly enhance the agent's understanding of side effects, fulfilling the transparency requirement.

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 two concise sentences, each packed with essential information. It front-loads the scope-dependent behavior and avoids any fluff. Every word contributes directly to clarifying the tool's function, making it highly efficient.

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 description fully explains the core scoping semantics and provides strong transparency, but it lacks details on parameter usage (beyond scope) and expected outcomes or errors. With no output schema, the agent might need additional guidance on return values or failure cases. The description is adequate for the primary decision but incomplete for a four-parameter destructive tool.

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?

With 0% schema description coverage, the description must compensate for parameter semantics. It explains the `scope` enum (turn vs. queue) in context, but it completely omits `taskId`, `sinceCursor`, and `expectedRevision`. While `taskId` might be inferred from the name, the cursor and revision parameters remain unexplained, leaving gaps in parameter understanding.

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 action (cancel/remove) and the resource (active root turn vs. queued items) with precise scoping. It distinguishes between two modes (`turn` and `queue`) and their distinct effects, making the tool's purpose unambiguous and differentiating it from any sibling tool.

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 provides explicit context for when to use each scope: `turn` cancels only the active turn, `queue` removes items from the queue. While it doesn't state alternatives or exclusions, the behavior is clearly tied to the two modes, giving the agent enough guidance to choose appropriately. It could be improved by mentioning when cancellation is not applicable, but it's generally clear.

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