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

Annotations already cover the safety profile (readOnlyHint=false, destructiveHint=false, idempotentHint=false), so the description only needs to add context above that. It does add one genuinely useful behavior — an empty value deletes the entry — but says nothing about whether an existing key is silently overwritten, whether it works on stopped runs, or what the call returns. Note the mild tension between 'empty value removes it' and destructiveHint=false.

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?

Two short sentences, zero filler, with the set semantics first and the edge case second. Nothing is wasted or buried.

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 3-required-param mutation with annotations and no output schema, this is only just enough. The agent still cannot tell whether the call overwrites existing values, what happens on a completed run, or what the response body contains beyond the schema-declared inputs.

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 coverage is 100%, so all three parameters (key, run, value) are already documented in the schema. The description adds only one extra semantic — value='' means delete — and does not explain accepted value encodings or whether key namespaces collide with other run data. Baseline 3 is appropriate when the schema carries the load.

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 concrete effect — storing a named value on a run ("a reading or a number") — and adds the important twist that an empty value removes it. It is clear what the tool does, though it never contrasts itself with siblings like add_note or add_step that also attach data to a run.

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?

The only usage-ish signal is the empty-value deletion rule. There is no statement of when to prefer set_value over add_note/add_step, no prerequisites (must the run be active? paused?), and no guidance on overwrite behavior for an existing key.

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