Forget a user's cached check-ins

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

Annotations already mark the tool destructive and idempotent, and the description substantially expands on that with essential behavioral details: it asks for confirmation first, shows a preview with username and row count, and explains the two-step confirmToken fallback for clients without elicitation. It also discloses that shared beer metadata survives and that the operation is reversible via a later sync, which is valuable beyond the structured annotations.

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 longer than typical but every sentence earns its place, front-loading the core destructive effect before moving into confirmation behavior and token handling. It is slightly dense, but for a destructive cache-clearing tool with a two-step confirmation flow, the detail is justified and well organized.

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?

Given the tool's destructive nature, the absence of an output schema, and the unusual two-step confirmation flow, the description is remarkably complete. An agent knows what will be deleted, what will be preserved, what the preview shows, how to handle both confirmation modes, and what happens on a later sync. Nothing essential for correct invocation is missing.

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 already covers both parameters at 100%, including the meaning of omitting username. The description adds important operational semantics beyond the schema: confirmToken must come from phase-1, only after explicit user approval, never reused, and passed with identical arguments. This turns a bare token parameter into an understandable workflow.

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 specific verb ('Delete') and resource ('everything the LOCAL cache holds for one user'), and enumerates exactly what is covered: cached check-ins, distinct-beers list, and sync state. It clearly distinguishes local-cache behavior from anything on Untappd, so an agent can tell it apart from remote delete tools like untappd_delete_checkin. The scope boundary ('Shared beer metadata is kept') sharpens the purpose further.

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 description gives concrete context for when to call it ('after syncing a friend you no longer want a history of') and clarifies that only the local cache is affected while a later sync can re-fetch data. It does not explicitly name sibling alternatives such as untappd_cache_query or untappd_delete_checkin, but the local-vs-remote distinction and 'nothing on Untappd changes' strongly imply when this tool is and is not appropriate.

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