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

Annotations already flag destructive=true and idempotent=true, but the description adds substantial behavioral context beyond them: the cascade refusal rule, the strong 'never deletes a memory the user wrote' guarantee with its three-part precondition for deleteBackingMeta, and the dryRun-by-default safety behavior. That is exactly the kind of context annotations cannot express.

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?

Purpose is front-loaded in the first clause and every subsequent sentence carries an actual rule rather than filler. It is a single dense paragraph with nested conditions that read a little run-on, but no sentence is wasted.

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, no-output-schema tool this covers what an agent needs: what is removed (node plus links), the refusal conditions, the memory-preservation guarantee, and the dryRun default. Nothing material is left unstated.

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 coverage is 100%, so the schema already documents all four parameters and the baseline is 3. The description goes further by explaining the interaction semantics: cascade removes 'the whole section,' deleteBackingMeta fires only when the memory holds nothing but a title/description and has no other referrers, and dryRun defaults to a non-committing report.

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?

States a specific verb and resource ('Remove an item or section from a bundle') and defines a bundle inline as 'an ordered outline of memories for a book or project,' with the important scope note that all links to/from the node go too. It stops short of naming a sibling alternative (e.g. muninn_bundle_nodes_remove or muninn_memory_delete), so an agent still has to infer the single-vs-multiple split.

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?

Gives clear conditional guidance: it refuses a section that still has items unless cascade=true, and it deletes the backing memory only under three explicit conditions. However it never points to the alternative tool for bulk removals or explains the memory-vs-node deletion boundary against muninn_memory_delete.

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