Experiment Artifacts

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

Annotations already declare readOnlyHint=true, so the safety profile is covered. The description adds genuinely useful behavioral context by disclosing the "bounded byte cursor" pagination mechanic, but it says nothing about error handling for missing paths, how cursor and offset interact, or response shape expectations.

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?

A single front-loaded sentence that conveys the tool's two modes with zero filler. Well-sized for the information it chooses to deliver.

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?

Although an output schema exists and annotations cover read safety, five parameters with 0% schema coverage and two distinct operating modes are not adequately explained by one sentence. An agent can guess at path/cursor but cannot confidently use limit or offset, making the definition incomplete for the tool's actual complexity.

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 5 parameters, so the description must carry the burden. It only hints at path (read one artifact) and cursor (bounded byte cursor); run_id, limit, and offset receive no explanation, and the critical relationship between cursor and offset is undefined.

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 two specific modes with a clear resource: listing a run's files and reading a single artifact. The verb+resource is unambiguous and the dual-mode framing helps an agent understand the tool's identity, though it does not explicitly contrast itself against siblings like object_get or experiment_rollouts.

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?

Usage is only implied: the phrase "or read one artifact" suggests that supplying a path switches to single-artifact read mode, but there is no explicit when-to-use, when-not-to-use, or named alternative. Adequate but leaves mode selection to inference.

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