agent-learn

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

Annotations already communicate idempotence, non-destructiveness, and a closed world, so the bar is lower. The description adds that this is a thin wrapper that forwards to learn with a fixed sectionType, and it discloses the deprecation lifecycle, which annotations do not. However, it does not explain what the learn operation actually does with the content, such as storage, merging, overlap, or response behavior.

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 only three sentences and every sentence contributes: the shorthand relationship, the content semantics, and the deprecation. It is front-loaded with the core identity and ends with the lifecycle warning. Slight room for a usage guidance sentence, but as written there is no fluff.

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 no output schema, so the description should help fill what happens and what the caller receives. It does not mention the effect of learning persona text, whether content replaces or edits existing agent knowledge, what the response looks like, or how the sectionType agent maps into the learn workflow. It is only complete for a caller who already deeply understands the learn tool.

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 schema has 0% semantic description coverage, so the prose must compensate for the 7 parameters. It only clarifies that content is the persona text; no other parameters (tags, type, docid, title, replace, description) are explained. It also does not clarify that sectionType is hardwired and whether a caller can override the type property.

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 identifies the operation as 'shorthand for learn with sectionType: agent' and states that content holds the persona text, which is specific enough to say what it does and how it differs from the base learn tool. It is clear about the resource (agent/persona) and the action (learn-like behavior), though it still relies on the caller already knowing what 'learn' itself does.

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 deprecation warning says 'Deprecated; removal in 2.0.0.' which is an unambiguous when-not-to-use signal. By calling itself a shorthand for learn, the alternative is implicitly available, but there is no explicit 'use learn instead' statement or discussion of when to choose this tool over related agents like agent-search or skill-learn.

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