Machine Status

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

With no annotations, the description carries the full burden and does well: it discloses admission semantics (whether a new GPU or CPU job may start), per-job memory accounting with OVER flags when past declared limits, and that heavy operations self-refuse rather than silently queueing. It never explicitly says the call is read-only/no side effects, which is the remaining gap.

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?

It is one long run-on sentence with nested parentheses and an awkward front-load ('The shared machine before anything heavy'), which makes the admission-control core harder to extract. Nearly every clause is substantive, so the problem is structure rather than padding.

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?

An output schema exists, so return values need no explanation, and the behavioral side (admission, OVER accounting, self-checking ops) is covered. However the lone `project` parameter is left completely undefined and the wrapper-command guidance is compressed into a single clause, leaving an agent to guess how to actually invoke the run subcommand correctly.

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% for the single `project` parameter, and the description never mentions it, so an agent cannot tell whether it scopes the status report to a project or filters running jobs. The description's extensive detail about what gets reported partially compensates for the low coverage, but the one parameter remains undocumented in both places.

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 resource domain and enumerates exactly what it reports: GPU heat/throttling, CPU, free memory, running heavy jobs, admission for new GPU/CPU jobs, free commit and disk. That distinguishes it from siblings like machine_disk, stage_status, or live_status, though no sibling is named explicitly.

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?

Gives concrete triggering conditions ('before anything heavy: a render over a minute, separate, mimic_measure, a fit, Blender, whisper') and notes that heavy ops already self-check and refuse with a reason, which implicitly bounds when a manual call is unnecessary. It also routes a related need (running external commands) to the `python -m ismail.machine run` form, though it stops short of naming when NOT to call the tool outright.

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