Skip to main content
Glama
蓝汐 · 潮汐认知图谱

蓝汐 · 潮汐认知图谱

by A16Nights

cg_maintain

Maintain a cognitive graph by processing maintenance queues, promoting scenarios, touching access, setting importance, consolidating nodes manually or with AI, backfilling embeddings, and rebuilding indexes.

Instructions

潮汐维护:查看维护队列(queue),晋升情景(promote),强化访问(touch),批量修改/确认重要度(set_importance),手动或 AI 智能凝结(consolidate / ai_consolidate),补全向量(backfill_embeddings),或重建索引(rebuild)。

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "cg_maintain"
  }
}

Input Schema

NameRequiredDescriptionDefault
idNopromote/touch/set_importance 时的目标节点 ID
opNo默认 queue
idsNoset_importance 批量目标节点 ID 列表
limitNobackfill_embeddings 批量上限,默认 50
titleNoconsolidate 新节点标题
commitNoai_consolidate 时是否直接写入落盘,默认 false 仅预览草稿
confirmNoset_importance 时是否同时确认保护(将 hint 转为 confirmed)
summaryNoconsolidate 提炼后的摘要正文
overrideNoset_importance 降低受保护节点重要度时传 true
sourceIdsNoconsolidate / ai_consolidate 的源节点 ID 列表
importanceNoset_importance 要设置的重要度 0~1
confirmedByNo确认来源标识,默认 'user'
targetLayerNopromote/consolidate 目标层
archiveSourcesNo凝结后是否归档非保护源节点

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed7 schema fields changedv1.1.1
    • addedInput schema / properties / confirm
      Added value: +{
      +  "description": "set_importance 时是否同时确认保护(将 hint 转为 confirmed)",
      +  "type": "boolean"
      +}
    • addedInput schema / properties / confirmedBy
      Added value: +{
      +  "description": "确认来源标识,默认 'user'",
      +  "type": "string"
      +}
    • changedInput schema / properties / id / description
      Previous value: -"promote/touch 时的目标节点 ID"New value: +"promote/touch/set_importance 时的目标节点 ID"
    • addedInput schema / properties / ids
      Added value: +{
      +  "description": "set_importance 批量目标节点 ID 列表",
      +  "items": {
      +    "type": "string"
      +  },
      +  "type": "array"
      +}
    • addedInput schema / properties / importance
      Added value: +{
      +  "description": "set_importance 要设置的重要度 0~1",
      +  "type": "number"
      +}
    • changedInput schema / properties / op / enum
      Previous value: -[
      -  "queue",
      -  "promote",
      -  "touch",
      -  "consolidate",
      -  "ai_consolidate",
      -  "backfill_embeddings",
      -  "rebuild"
      -]New value: +[
      +  "queue",
      +  "promote",
      +  "touch",
      +  "set_importance",
      +  "consolidate",
      +  "ai_consolidate",
      +  "backfill_embeddings",
      +  "rebuild"
      +]
    • addedInput schema / properties / override
      Added value: +{
      +  "description": "set_importance 降低受保护节点重要度时传 true",
      +  "type": "boolean"
      +}
  2. First observedv1.0.0

TDQS

C2.6/5.0
Behavior2/5

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

With no annotations and 14 parameters, the description carries the full behavioral burden yet discloses nothing about consequences. Destructive or consequential paths — rebuild (rebuild index), override (lowering importance of protected nodes), ai_consolidate with commit=true, archiveSources — get no warning, and there is no mention of permissions, reversibility, or rate/scale concerns.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness3/5

Is the description appropriately sized, front-loaded, and free of redundancy?

One front-loaded sentence with the resource stated first, which is efficient, but much of the length is spent restating the enum members rather than adding decision-relevant information. It is compact yet low-density in value.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness3/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a multiplexed 14-parameter maintenance tool with no annotations and no output schema, the definition is thin: it lists modes but never explains outcomes, safety, or the workflow linking ops (e.g. queue → promote → consolidate). The rich schema partially compensates, keeping this at minimum-viable rather than inadequate.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so every parameter including per-op applicability, defaults, and the enum is already documented in structured data. The description adds only one-line glosses for the op values and no extra meaning for id, ids, override, commit, or archiveSources; baseline 3 is correct.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose3/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description names the resource (潮汐 maintenance) and enumerates eight specific operations with short glosses (晋升情景, 强化访问, 重建索引), so an agent can tell roughly what the tool does. However, the enumeration largely restates the `op` enum values, and nothing distinguishes this maintenance surface from siblings like cg_update or cg_graph.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines2/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

There is no guidance on when to choose this tool over cg_update, cg_link, or cg_reflect, and no guidance on when to pick one op over another beyond the gloss. The reader must infer all selection logic from the enum alone.

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