run_destructive_command

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

There are no annotations, so the description carries the full burden of behavioral disclosure. It states that the command is risky, may break things, requires restating the risk in the agent's own words, requires a fresh explicit confirmation in the next message, and must not be used when no human is present. This is thorough transparency for a destructive operation.

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 by adding a safety constraint or workflow step. The example clarifies what does NOT count as sufficient confirmation, and the core precondition is front-loaded. It could be slightly more polished, but the length is justified for a destructive tool.

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?

This is a high-complexity, high-stakes tool with no output schema and no annotations, yet the description covers all essential context: the exact precondition, the human-confirmation protocol, the automated-run exclusion, the cwd routing rule, and the list_allowed_roots lookup. An agent has everything needed to decide when and how to invoke it safely.

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?

The input schema already documents all 5 parameters with 100% coverage, so the baseline is 3. The description adds meaningful usage-level semantics: when to pass cwd (target project other than primary) and how to get allowed roots via list_allowed_roots. It also clarifies that confirm:true must not be set automatically in reaction to rejection, which goes beyond the schema's generic confirmation description.

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 positions this as the dangerous fallback to run_safe_command: it runs commands that were rejected as risky, but only after explicit user confirmation. It distinguishes itself from the sibling run_safe_command through the rejection+confirmation precondition, even though it never uses a simple 'runs a command' verb phrase.

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 usage condition is explicit and complete: use only after run_safe_command rejected the command AND the user explicitly confirmed in chat. It also gives when-not-to-use (automated runs without a human), warns against setting confirm:true automatically, and points to alternatives like list_allowed_roots for cwd validation.

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