learn_flow

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

No annotations are provided, so the description carries the full behavioral burden and falls well short: it doesn't say whether this mutates the rule/claim store, how long a learn run takes, what 'oracle budget' means operationally, or what happens when the budget is exhausted. The one hint of cost-awareness is not actionable.

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?

It is a single front-loaded sentence with no padding, which is structurally sound. However, the space that is used is spent on unexplained abbreviations and a budget concept that maps to no input, so brevity comes at the cost of usefulness rather than from efficiency.

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?

An output schema exists, so return values needn't be described, but for a 5-parameter, no-annotation, presumably long-running learning operation the description leaves out mutation side effects, budget semantics, algorithm choices, and parameter meanings. It is not complete enough for an agent to invoke this correctly on the first attempt.

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 compensate and does not: 'feature' is only implicitly referenced, 'KV or L#' loosely maps to the 'algorithm' parameter without naming it or listing valid values, and 'alphabet', 'walk_len', and 'walks_per_state' are entirely unaddressed — including the required 'alphabet' array.

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 verb+resource ('learn a state machine for a feature') is identifiable, but the qualifiers 'KV or L#' and 'explicit oracle budget' are unexplained jargon and the budget concept has no corresponding parameter. An agent can guess the general purpose but cannot confidently predict what the tool will do with its inputs.

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 statement of when to use this versus the many sibling tools (check_rules, coverage, discover_ui, suggest_probes, etc.), nor any prerequisite or follow-up guidance. The phrase 'under an explicit oracle budget' hints at a cost constraint but never says when that constraint applies or how to satisfy it.

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