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

Annotations already cover read-only, idempotent, non-destructive and closed-world, so the bar is lower, yet the description still adds concrete failure behavior: an id from another graph 404s instead of resolving, and ids are not portable. It also gives the agent a decision rule for value_conflicts (state the disagreement rather than one value). It does not discuss pagination or size limits for large hub objects, which keeps it off a 5.

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 what is returned, then the citation rule, then the conflict rule, then input provenance, then exclusions — a sensible priority order with no filler. It runs long and includes agent-behavior instructions ('Cite those when you state a fact') that could be trimmed, but every block carries usable information.

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 two-param read whose output schema already defines the return shape, this covers the remaining gaps: what the fields mean semantically, how to source values, what a cross-graph miss looks like, and which sibling to use instead. Nothing an agent needs in order to invoke it correctly 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?

Schema description coverage is 0%, so the description must carry the two parameters and it does: `memory` is identified as the graph slug from list_my_memories, and `object_id` as coming from semantic_search, list_objects or recall, plus the crucial constraint that ids are graph-scoped and not portable. It stops short of stating the id format or slug shape, which the schema leaves as bare string/integer.

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?

Opens with a specific verb and scoped resource: everything the graph knows about ONE object, then enumerates the payload (properties, grouped incoming/outgoing relationships, actions, approved owners, evidence, value_conflicts). That enumeration makes it immediately distinguishable from list_objects, recall, or semantic_search without opening any schema.

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?

Explicitly routes the agent away from this tool: 'Reach for something else when' names semantic_search with expand=1, get_paths, and get_change_impact, each paired with the intent that selects it. It also states where the inputs come from (list_my_memories, semantic_search, list_objects, recall), so the agent knows the prerequisites before calling.

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