Ranks missing evidence by expected information gain and hypothesis probability shift, so you can prioritize interview questions and data requests.
아직 확보하지 않은 증거 중 가설 판정을 가장 크게 바꿀 증거를 순위대로 돌려줍니다 (기대 정보이득, 불리·유리 시 대상 가설 확률). 인터뷰 질문·자료 요청 목록으로 활용합니다.
| Name | Required | Description | Default |
|---|---|---|---|
| target | No | ||
| pattern | Yes | ||
| benchmark | No | ||
| stage_gap | No | ||
| observations | No |
Changes observed during successful MCP inspections.
v0.1.0Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations exist, so the description carries the full burden. It usefully discloses return semantics (ranked evidence, expected information gain, hypothesis probabilities under favorable/unfavorable outcomes) in place of a missing output schema, but it says nothing about required preconditions, whether the call is a pure read, or how results change with partial observations.
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 sentences, no filler, with the core purpose front-loaded and the practical use case trailing as secondary detail. The density is high given the length, though the parenthetical detail about probabilities is arguably more than the selection decision needs.
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 compensates reasonably for the absent output schema by describing what is returned, but against 5 parameters at 0% coverage, no annotations, and a math-heavy domain (information gain, benchmark comparison), it leaves the agent unable to supply inputs or confirm prerequisites. It is not adequate for a tool of this complexity.
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?
Five parameters with 0% schema description coverage, and the description names none of them. '대상 가설' loosely gestures at the `target` field, but `pattern` (required), `benchmark`, `stage_gap`, and `observations` are entirely unexplained in both schema and prose, leaving an agent unable to populate them correctly.
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 specific verb and resource: it returns un-obtained evidence ranked by impact on hypothesis assessment, and it even names the ranking criterion (expected information gain) and the probability framing. That is far more than a tautology, though it never names which sibling it is not (e.g. assess_hypotheses or recommend_actions), so sibling differentiation must be inferred.
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?
It gives one concrete downstream use — turning the output into interview questions or a data-request list — which implies the 'value of information collection' context. However, it never states when to call this rather than assess_hypotheses or recommend_actions, nor any prerequisite such as needing existing observations for the calculation to be meaningful.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.