Store a piece of information worth remembering long-term into a memory bank; repeated content reinforces existing memories instead of duplicating them.
把一条值得长期记住的信息写入记忆库（分主次存储，重复内容会强化而非重复）。
| Name | Required | Description | Default |
|---|---|---|---|
| kind | No | 记忆类型，可省略，自动判断 | |
| entity | No | 涉及的人或物，可省略 | |
| content | Yes | 要记住的内容，一句自然语言 | |
| profile | No | 角色/用户分区，默认 default | |
| importance | No | 重要度 0~1，可省略，自动估算 | |
| trigger_terms | No | 唤醒线索词，被提到时更容易想起，可省略 |
Changes observed during successful MCP inspections.
v0.4.0Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description must carry behavioral burden; it does add real value by disclosing that storage is split into primary/secondary tiers and that duplicate content reinforces rather than duplicates. However, it omits permissions, whether writes are reversible, and what the agent gets back after storing.
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?
A single front-loaded sentence that states the purpose first and the non-obvious dedup/reinforcement behavior second. No filler or redundant restatement of the tool name.
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 six-parameter write tool with full schema coverage and no output schema, the description covers the essential semantics including the surprising merge-on-duplicate behavior. Only the lack of any statement about side effects on near-duplicate or conflicting memories keeps it from a 5.
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 coverage is 100%, so the schema already documents all six parameters including the kind enum and importance range. The description adds only the duplicate-reinforcement behavior tied to content, which does not meaningfully extend per-parameter meaning, making the baseline 3 appropriate.
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+resource (写入记忆库 / write into memory store) and scopes it to information worth remembering long-term, which distinguishes it from read-oriented siblings like memory_recall and memory_list. It does not explicitly differentiate from the closer sibling memory_update, so it falls short of a 5.
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?
Implies when to use it ('值得长期记住的信息'), giving a usable criterion, but never names alternatives such as memory_update for existing entries or memory_pin for emphasis. The duplicate-handling note hints at usage but no explicit when-not guidance is present.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.