Record evidence-backed study claims as facts or inferences, or flag unknowns, while noting supporting evidence and limitations for traceable findings.
记录研究者的材料事实 fact、分析推断 inference 或待确认事项 unknown。fact/inference 必须引用本研究证据；inference 必须说明局限。引用检查不证明结论正确。
| Name | Required | Description | Default |
|---|---|---|---|
| text | Yes | ||
| type | Yes | ||
| study_id | Yes | 研究 ID，由 research_create 返回 | |
| limitation | Yes | ||
| evidence_ids | Yes |
Changes observed during successful MCP inspections.
v0.4.0Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare a non-readonly, non-destructive, non-open-world write, so safety is covered. The description adds genuine behavioral value in the form of validation semantics (citation requirement, limitation requirement) and the caveat that a passing citation check does not validate the conclusion. It stops short of saying what happens on failure or whether claims are deduplicated.
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?
Three short, dense sentences with the core purpose front-loaded and constraints following. No filler, though the terse phrasing borders on cryptic.
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 5-required-param write tool with no output schema and thin schema descriptions, it covers purpose and the key type/evidence/limitation rules adequately. It omits any interaction with the update/evidence siblings and says nothing about failure or duplicate handling.
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 only 20% (only study_id is documented), so the description must carry weight. It does define the meaning of the type values and ties evidence_ids and limitation to specific type values, which is useful, but leaves text and the study_id semantics largely to the schema.
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+resource (record researcher claims) and enumerates the three claim types it accepts: fact, inference, unknown. That lets an agent understand the tool's output domain without opening the schema. It does not, however, distinguish this from close siblings such as research_add_evidence or research_add_verification.
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 real conditional rules — fact/inference must cite this study's evidence, and inference must state a limitation — which steers correct invocation. But there is no explicit when-to-use-vs-alternative guidance and no mention of sibling tools like research_add_evidence, so routing must be inferred.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.