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

With no annotations provided, the description carries full behavioral burden and does so well: it discloses the production owner-approval requirement, the 30s timeout, the 20 KB output cap, the read-only code with writable /tmp, and the pending_approval flow with polling via approval.status. This is exactly the context an agent needs beyond the schema.

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?

Front-loaded with the critical production-approval caveat, then the core action, then limits, then the approval flow. Every sentence earns its place. Slightly dense but well organized; minor room for trimming the example list.

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 no output schema and no annotations, the description provides a remarkably complete picture: execution context, limits, write constraints, and the approval/polling workflow with the correct sibling reference (approval.status). An agent can invoke this correctly without further info.

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 coverage is 50%: 'env' and 'project' are documented in the schema. The description adds no explicit semantics for 'command' or 'timeout_seconds' (which are undocumented in both). It slightly clarifies 'command' via examples like 'node -e ...' and 'ls', but 'timeout_seconds' remains unexplained. Baseline 3 is appropriate given the partial coverage.

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+resource ('run a shell command inside the environment's running sandbox') with the exact execution context (same code, env vars, network, limits as the app). The description distinguishes this from siblings like code.read or db.query by scoping it to shell execution inside the running sandbox.

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?

Clearly implies when to use it: to inspect or execute within the live sandbox, with concrete examples ('node -e ...', 'ls', 'env | sort'). It also notes the approval condition for production and directs to approval.status. However, it doesn't explicitly rule out alternatives or say when NOT to use it (e.g., vs. code.read or logs.query).

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