Purge Knowledge Base

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

Beyond the annotations (readOnlyHint=false, destructiveHint=true), the description discloses irreversible deletion, confirm:true requirement, error cases for ambiguous/invalid scopes, deprecated bare-confirm behavior, and the return summary. This gives the agent a rich behavioral contract without contradicting the 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 organized with clear sections (DESTRUCTIVE, WHEN, WHEN NOT, SCOPES, CONTRACT, RETURNS, EXAMPLE), making it scannable and front-loaded with the most critical warning. Every line conveys necessary behavioral or usage information with no filler.

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 highly destructive tool with no output schema, the description fully covers return values, error conditions, deprecated paths, and scope-specific effects. It addresses all three parameters and even includes examples, making it complete for an agent to select and invoke the tool correctly.

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?

Although the input schema already fully documents each parameter, the description adds critical meaning: exactly one scope must be passed, sessionId cannot be combined with scope:'project', scope:'session' requires sessionId, and confirm:false cancels. It also provides two valid example invocations, which materially improves correct usage.

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 'DESTRUCTIVE: permanently delete indexed content. Cannot be undone,' which clearly identifies the action, resource, and severity. It also distinguishes the two scopes (session vs project), making the tool's purpose unmistakable relative to its siblings.

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 provides explicit WHEN and WHEN NOT sections, including concrete user phrasing triggers and instructions to ask for a scope if ambiguous. It even recommends an alternative tool (ctx_stats) for memory/performance concerns, giving clear guidance on when not to purge.

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