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

Beyond annotations it discloses the auth requirement (project grant executor.run), a quota rule (each env secret consumes one secrets.use), a hard security constraint (secrets only via env as vault://service/key, never in the command), and state persistence (/work survives between calls while the VM lives). These are non-obvious operational facts an agent cannot infer from readOnlyHint/destructiveHint alone.

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 execution target and the server-safety contrast, then moves through auth, secrets, persistence, and the wait/poll decision. It is dense but nearly every clause carries operational weight; the parenthetical aside in the 'wait' sentence is the only slightly tangled construction.

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 9-parameter, nested-object, no-output-schema tool the description covers the highest-risk concerns: auth, secret handling, VM locality, working-directory persistence, and the synchronous-versus-async return path. It remains incomplete on four parameters (network, timeout, workspace, result_offset) whose semantics are left entirely to the bare schema.

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 0% across 9 parameters, so the description carries the full burden. It meaningfully explains 'wait' (range, semantics, why it is cheaper than polling), 'env' (vault references and quota), and implies the command/argv and project_id use case, but says nothing about 'network', 'timeout', 'workspace', or 'result_offset'.

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 opening sentence names a specific verb and resource: 'Run your own command in a disposable sandbox (bubblewrap) on the account executor VM, never on the Fileaify server.' It pins down both what executes and where, which an agent can act on 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?

It routes the agent explicitly to executor_sandbox ('Read executor_sandbox first: it lists ready-made recipes') and explains when to use 'wait' versus returning a job_id polled with executor_job. The gap is that sibling executor_ssh_run also runs commands remotely, and the description never says when to pick one over the other.

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