Start Embeddings

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

No annotations are provided, so the description carries the full burden. It does disclose the semantics of four behavior-changing flags (profiling, no_cache, legacy, use_last_checkpoint), which is real value, but it omits key traits for a compute job: that it is long-running, that it writes embeddings to disk, and whether existing outputs are overwritten.

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 lead sentence front-loads the purpose and scope, followed by terse one-line flag definitions. Structure is good and there is little waste, though the flag block could be better tied to the parameter names an agent must supply.

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 15-parameter, no-annotation tool with 0% schema coverage, the description is thin: it omits prerequisites, expected runtime/compute implications, output locations, and most parameter meanings. An output schema exists, so return values need not be explained, but the invocation-side gaps are significant.

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 15 parameters, so the description must compensate and only partially does. It explains profiling, no_cache, legacy, and use_last_checkpoint, but leaves 11 parameters undocumented, including the required models_path and datasets_root and the meaning of cpu, output, regions, subjects, masks_version, and cortical_version.

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 states a specific verb and resource ('compute sulcal embeddings') and pins down scope ('across all 56 model folds (28 regions x 2 hemispheres)') plus its place in the pipeline ('Stage 4'). It does not explicitly contrast itself with siblings like start_training or start_cortical_tiles, so it falls short of full differentiation.

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 when-to-use or when-not-to-use guidance and no mention of the alternatives in the sibling set. The only implied context is the 'Stage 4' label, which hints at pipeline ordering but does not state prerequisites (e.g., that models must already be trained) or how it relates to start_pipeline.

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