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

Annotations already cover the safety profile (readOnly, idempotent, non-destructive), and the description goes well beyond them: it explains the 'status' semantics ('unapproved' rows are human proposals), that review_queue counts pending approvals, and that conflicts must be surfaced rather than resolved. That is exactly the interpretive context an agent needs to avoid inventing facts.

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?

Front-loaded with the imperative 'CALL THIS FIRST', then organized into return contents, parameter meanings, and exclusions. Every sentence carries load; length is justified by the breadth of the composite read.

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 description covers trigger, contents, parameter semantics, and interpretation guidance for the returned fields. Although an output schema exists, the description still provides the interpretive layer (status handling, conflicts) that raw schema alone would not convey.

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%, so the description carries the full burden and does so for all three params: memory is the graph slug from list_my_memories, topic additionally runs hybrid search and inlines hits, and limit bounds the recent list. It even explains the design rationale for topic being a single call rather than two.

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 ('composite read that answers what a cold agent actually needs') and enumerates exactly what it returns: classes and counts, recent activity, conflicts, review queue, and session checkpoint. This clearly distinguishes it from the sibling reads an agent might otherwise pick.

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?

Explicit 'CALL THIS FIRST, before working in a graph' sets the when, and the closing line names when not to use it plus the specific alternatives: semantic_search for a topic, get_object for one thing, list_objects for a population. Nothing is left to inference.

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