kernel_brain_decay

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

With no annotations, the description carries the burden and does disclose the important behavior: this mutates every learned weight in place and its effect is to fade stale boosts/penalties. It never states that the change is irreversible, whether it applies globally or per-scope, or what happens on repeated calls — material gaps for a destructive maintenance operation.

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 tight sentences with zero filler; the core mechanism is front-loaded ahead of the scheduling note. Slightly clipped, but nothing is wasted.

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 one-parameter maintenance mutation with an output schema available, the description covers what it does and roughly when. It omits mutation safety, reversibility, scope of effect, and rate bounds — gaps that matter given zero annotation coverage.

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 0% for the single `rate` parameter, so the description must compensate. 'Pull ... a fraction back toward the 1.0 baseline' effectively explains that rate is the fraction of the gap to baseline that is closed, which is meaningful beyond the bare `number, default 0.1` — but the valid range (0–1) and behavior at extremes are unstated.

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 specific verb+resource: pulling learned weights back toward a 1.0 baseline so stale boosts/penalties fade, prefixed with the 'Maintenance' category. That's clearly distinguishable from siblings like kernel_brain_ask, recall, or feedback, though no sibling is named explicitly.

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?

'Run periodically' implies a scheduled-maintenance trigger, which is real when-to-use guidance. It gives no when-not condition (e.g. don't run mid-experiment), no cadence, and no alternatives — an agent must infer the rest.

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