profile_job

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

Annotations already establish readOnlyHint=false and destructiveHint=false, so the safety profile is largely covered. The description adds real context: it uses NVIDIA Nsight Systems and windows the capture to avoid an enormous trace. It still omits job-submission behavior, resource consumption, and what the confirm parameter changes, so it is only moderately transparent for a mutating tool.

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?

Three sentences, front-loaded with the core action and followed by the reason and the handoff to profile_report. The phrase 'au lieu de laisser deviner pourquoi un code est lent' is persuasive padding, but overall the description is efficient and easy to scan.

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 13-parameter mutating tool with an output schema, the description is too thin. The output schema means return values need not be explained, but the description should still cover the profiled command, why windowing exists, and what the agent must provide or confirm. It leaves the parameter surface and operational requirements almost entirely unaddressed.

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?

There are 13 parameters with 0% schema description coverage, and the description explains none of them. The passing phrase 'La capture est fenetree' vaguely hints at the profiling-window parameters (profile_steps, warmup_steps, duration_seconds, delay_seconds), but it does not map to any argument, define units, defaults, or meaning. With this many undocumented parameters, the description fails to compensate for the schema gap.

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 concrete verb and resource: 'Profile un calcul GPU avec NVIDIA Nsight Systems et produit un rapport exploitable.' That is specific enough for an agent to know it launches a GPU profiling run. It does not differentiate itself from the sibling tool_profile, which appears to be a related profiling entry point, so it falls short of a 5.

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 a motivating context ('au lieu de laisser deviner pourquoi un code est lent') and a follow-up instruction ('Lis ensuite le resume avec profile_report'). However, it never says when to use this instead of alternatives such as tool_profile, nor does it state prerequisites or exclusions. Usage is implied rather than explicit.

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