process_kill

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

Annotations already declare destructiveHint=true, readOnlyHint=false, and idempotentHint=false, so the description correctly avoids restating them. It adds real value beyond annotations by disclosing the cascading child-tree kill default and the stateful side effect that the VM lock persists after the kill, naming vm_unlock as the required release step. These are behavioral facts not available in the structured fields.

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?

Two dense sentences with the core action front-loaded: the action, the default scope, and the lock/unlock consequence. Every clause earns its place, with no filler or redundant restatement of the tool name.

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 destructive 3-parameter tool with no output schema, the description covers the critical operational path: what gets terminated, the default tree behavior, and the mandatory vm_unlock follow-up. Minor gaps remain — how to discover process_id and what the call returns — but an agent can invoke it correctly with the information given, aided by the sibling process_list 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?

Schema description coverage is only 33% — process_id and tree are undocumented in the schema. The description partially compensates by mapping the 'tree' parameter to its default behavior ('by default, its discovered child tree'), but it never explains how to obtain process_id (e.g., via process_list) or the meaning of the tree flag beyond its default, leaving an agent to infer those semantics.

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?

States a specific verb and resource — 'Terminate a process' — and adds distinctive scope: 'by default, its discovered child tree.' This unambiguously separates it from sibling process tools (process_list, process_wait) and VM-level termination tools (qemu_vm_force_stop), so an agent can select it correctly without opening the schema.

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 implied context and a post-condition ('VM remains locked by this session; call vm_unlock when finished'), which tells the agent what to do after invoking it. However, it provides no explicit when-to-use guidance or exclusions versus alternatives such as shell_terminate for shell processes or qemu_vm_force_stop for whole-VM termination.

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