research_unknowns

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

No annotations are provided, so the description carries the full burden of behavioral disclosure. The description says 'Run targeted retrieval' but does not detail side effects, such as whether it modifies research state, requires prerequisites (e.g., research must be initialized), or if it performs destructive actions. It also doesn't explain what happens to the research data after retrieval, or if this is a read-only operation. The output schema is present but the description doesn't clarify the behavior beyond the basic action.

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 a single sentence, which is concise and to the point. It front-loads the action ('Run targeted retrieval') and the purpose ('for claims that still need work'). No wasted words. However, it is so vague that conciseness works against it, but structurally it is well-formed. A score of 4 reflects efficiency, but not necessarily clarity.

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?

Given the complexity of the tool—with nested objects, multiple retrieval settings, and a large set of sibling tools—the description is insufficient. It does not explain the expected input format for 'gaps', how the tool integrates with the research pipeline, or what the output packet contains. The output schema is present, but the description's lack of context leaves the agent uncertain about how to prepare inputs. With no annotations and 0% schema coverage, the description fails to provide the necessary context for correct invocation.

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%, meaning the description must compensate for the schema's lack of explanation. The description does not explain any parameters, but the parameter names and structure are somewhat self-explanatory (e.g., 'research_id' and 'gaps' with nested GapPlan). However, the nested objects like RetrievalSpec have default values and some descriptions for fields like domain_type, but this is in the schema, not the description. The description adds no value beyond what the schema implies, so a score of 3 is appropriate as a baseline for minimal compensation.

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 states a general verb phrase ('Run targeted retrieval') and a resource ('claims that still need work'), but it is vague. It does not specify what 'unknowns' means, how the tool relates to the research pipeline, or what 'compact review packet' contains. It lacks the specificity needed to distinguish it from similar tools like discovery_search or dispatch_parallel_subagents. The title 'research_unknowns' is not elaborated, so the purpose remains ambiguous.

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 description gives no explicit guidance on when to use this tool versus alternatives. It mentions 'claims that still need work' but does not explain how this tool fits in the research workflow (e.g., after assess_claims, before finalize_research). With many sibling tools, the lack of when-to-use or when-not-to-use guidance is a significant gap. No alternatives are named.

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