Logically delete selected observations for named entities from shared agent memory while retaining deletion history for compatibility and audit.
Compatibility tool backed by the same v2 SQLite store. New callers should prefer versioned native memory tools. Legacy deletion is logical and retained in history.
| Name | Required | Description | Default |
|---|---|---|---|
| deletions | Yes |
Changes observed during successful MCP inspections.
v0.2.0Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare destructiveHint=true and idempotentHint=false, so the description earns credit for adding that deletion is logical and retained in history — a crucial nuance for an otherwise destructive-looking operation. It does not add auth, limits, or output detail, but it meaningfully explains the destruction model.
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?
Three short sentences with no filler; the compatibility context, the preference rule, and the logical-deletion behavior are each stated once. Slightly better front-loading of what the tool does would help, but it is compact.
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 destructive legacy mutation tool with no output schema, the description covers deprecation status and deletion semantics but leaves the complex deletions payload entirely unexplained. An agent can decide whether to use it but not how to call it correctly without reading the 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?
Schema description coverage is 0% and the single parameter is a nested array of entityName plus observations with per-item limits. The description says nothing about this structure, so an agent must open the schema to learn what to pass.
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 frames the tool as a legacy compatibility path for deletion but never states plainly that it deletes observations for an entity; the resource is only inferred from the tool name. It does distinguish itself from newer native memory tools, which helps, but the core verb+resource statement is left implicit.
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?
"New callers should prefer versioned native memory tools" is a clear routing rule that steers agents away from this tool toward siblings like forget_memory. What is missing is a positive statement of when legacy callers should still use it, but the alternative is named explicitly.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.