voltage_lessons

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

Annotations declare readOnlyHint=true and the description aligns with that (no mutation implied). The description adds valuable behavioral context: lessons persist across sessions, are keyed by target, and include specific types of information (labels, HUD probes, timing). This goes beyond the annotation by describing persistence and content, which is useful for setting expectations about what the tool returns.

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 concise and front-loaded with the purpose. It uses bold for emphasis ('before writing a Playbook') and keeps each sentence purposeful. There is no filler or redundant explanation. The structure guides the reader from what the tool does, to when to use it, to how to filter results.

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?

The tool has an output schema (as indicated by the context), so return values are documented elsewhere. The description provides sufficient context for an agent to decide when to call it: it explains the purpose, when it is appropriate (before writing a Playbook for a previously driven target), and how to control scope with the target parameter. No critical information is missing, given the read-only annotation and output schema.

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?

The input schema has 0% description coverage, so the description must compensate. It clearly explains the `target` parameter (keyed by target, omit to see everything) and gives examples of valid values. However, it does not mention the `limit` parameter at all, leaving its semantics to inference from the default value of 30. This is a partial compensation but not complete for both parameters.

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 verb and resource: 'Recall what previous runs learned about driving something.' It then gives concrete examples of lesson content (labels, HUD probes, timing), which makes the tool's purpose unambiguous and distinct from any other sibling. The behavior is clearly scoped to recalling learned lessons, not a general-purpose query.

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?

It explicitly instructs when to use the tool: 'Call this **before writing a Playbook** for a target you have driven before.' It also explains the benefit (start from previous discoveries rather than rediscovering) and provides parameter guidance: 'Omit `target` to see everything recorded so far.' This gives an agent clear, actionable context for choosing this tool over alternatives like voltage_learn.

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