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

With only destructiveHint=false and openWorldHint=false in annotations, the description adds meaningful behavioral context: it discloses the decision criteria (relationships, sources, photo/LinkedIn, age), the rule that names alone never justify a merge, and the advisory nature ('decides nothing'). It does not discuss permissions or return payload, but the output schema exists to cover returns.

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?

The description is front-loaded with the core action and outcome options, followed by decision criteria and an important advisory caveat. Every sentence adds value, with no redundant restatement of schema.

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 the tool's purpose, decision logic, and non-mutating nature, which is substantial for a recommendation tool with an output schema. It lacks explicit guidance on how to obtain the single dedup pair or how it relates to scan/list duplicate tools, which would fully close the context gap.

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 50%, so the schema documents the UUID wrappers but leaves some parameters under-described. The description adds meaning by elaborating the decision enum (MERGE/IGNORE/DEFER) and implying survivorId ('which record to keep'), but it does not cover companyId, candidateId, agentName, or note/reason explicitly, so compensation is only partial.

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 specifies a concrete verb (Recommend), resource (network duplicate), and scope (one dedup pair), and enumerates the three outcomes (MERGE, IGNORE, DEFER). This clearly distinguishes it from sibling tools like fundraising_list_network_duplicates and fundraising_scan_network_duplicates, which operate across many pairs.

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?

The phrase 'one dedup pair' establishes the context for use, implying this tool is applied after duplicates have been identified. It also clarifies that the 'recommendation decides nothing,' which is a useful usage boundary, but it does not explicitly name alternatives or state when not to use it.

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