get_relationship_graph

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

Annotations already declare readOnlyHint=true, so safety is covered. The description adds two useful facts beyond that: only 'active' relationships are returned (history excluded), and output is 'capped for bounded disclosure' — signalling possible truncation. However, it never says what the cap is, whether truncation is flagged, or what permissions gate access, so an agent cannot tell whether results are complete.

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?

Two tight sentences with the core behaviour front-loaded and the addressing options immediately after. Nothing is wasted, though the terseness comes partly from omitting information that mattered (depth, types, cap behaviour).

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 four-parameter read-only graph query with no output schema, the description leaves the essential shape of the response unstated — no indication of node/edge structure, what depth expands to, how capping manifests in results, or what the types filter accepts. An agent could call it, but could not reason about whether the returned graph is complete or filtered.

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, yet it explains only person_id and person (including the resolve_person provenance). It says nothing about depth despite the default of 2, and nothing about the types filter — the agent must guess whether types means relationship kinds, entity kinds, or something else.

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: 'Return active relationship structure around one person,' with the scope qualifier 'active' and the caveat 'capped for bounded disclosure.' An agent can tell this is a graph read, not a single-person lookup. It does not name the near neighbours (find_connection, explain_shared_connections, get_person_context), so sibling differentiation 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?

Provides addressing guidance ('Pass person_id from resolve_person, or person to resolve inline'), which is really an invocation hint rather than when-to-use guidance. It never says when this tool beats find_connection or explain_shared_connections, nor when a graph is inappropriate. Usage is implied by the name and the resolve_person dependency.

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