run_grid

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

No annotations are provided, so the description carries the full burden, and it delivers substantial context: it discloses the fallback ('Runs Radiance if present, else returns the commands'), the sky-detection heuristic (gensky/gendaylit primitives, skyfunc, hemispheres, basename hint), and even the failure consequence ('A leak here reports daylight as electric light... so an over-count is visible'). This is genuinely useful behavioral disclosure beyond the schema. It falls short of 5 only because it never addresses side effects such as file/octree writes or work_dir mutation.

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 front-loaded with purpose and uses parameter names as scannable markers, which is good structure. However, the sky_files paragraph is a dense wall of conditional logic that runs several sentences without a break, making it hard to parse. For a tool this complex some length is justified, but the text is not concise and could be tightened.

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 10 parameters, 0% schema coverage, and an output schema that covers return values, the description handles the hardest parameter (sky_files) extremely well but leaves run, quality, rad_params, and channel_type entirely undocumented. With no annotations and zero schema coverage, that is a real completeness gap — an agent cannot correctly configure those parameters from this description alone.

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%, so the description must compensate, and it does for some parameters: scene_sets (1 list vs 3 lists for 3/9-channel), points (row format [x,y,z,dx,dy,dz] or .pts path), export_csv (abs or relative path), and sky_files (detailed semantics tied to electric_lights). But roughly half the parameters — run, quality, rad_params, channel_type, and electric_lights' own semantics — receive no explanation at all. Partial compensation for a large 10-parameter surface.

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 opens with a specific, informative statement: 'Point-in-time grid HCL metrics via Radiance (oconv -> rtrace -I per channel -> recombine)'. This names a clear verb+resource and even spells out the pipeline, which implicitly distinguishes it from siblings like run_annual and run_image. It is not a tautology and not vague, though it never explicitly names a sibling for contrast, so it stops 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?

There is no guidance on when to choose this tool over alternatives such as run_annual, daylight_factor, or run_image. The description explains mechanics (how Radiance is invoked, fallback behavior) but never states the use-case boundary or exclusions. An agent gets no help deciding whether point-in-time HCL grid metrics is the right call versus an annual or imaging sibling.

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