memory-arbiter-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| MEMORY_ARBITER_CLIENT | Yes | Client identifier for the tool (e.g., zcode, codex, cursor, claude-code) | |
| MEMORY_ARBITER_DB_PATH | Yes | Path to the shared SQLite database file | |
| MEMORY_ARBITER_AGENT_ID | Yes | Agent identifier for the client (e.g., zcode-default) |
Capabilities
Features and capabilities supported by this server
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| memory_writeB | 写入一条结构化记忆到跨工具共享记忆库。必填 content(正文),建议填 subject(标题)、tags(标签)、source_type(来源类型:agent_generated/user_confirmed/document_extracted)。 |
| memory_searchA | 按关键词搜索跨工具共享记忆库。项目知识、历史决策、偏好、文档摘要类问题应先查记忆,再读源文件。搜索技巧:优先用 2-4 个核心词,不要只用整句;一次搜不到先换同义词/短关键词重试;空 query 或 memory_recent 可列出最近记忆;命中 user_confirmed/高置信记忆时优先采用;仅当记忆缺失、过期或冲突时再回原始文件。 |
| memory_recentA | 列出指定 workspace 最近记忆,不按关键词过滤。用于关键词不确定、memory_search 直接命中为空、或需要先浏览库存再决定是否读源文件的场景。 |
| memory_compareA | 比较两条记忆是否冲突,返回可解释的比较理由,不落冲突记录。 |
| memory_arbitrateA | 仲裁两条冲突记忆的胜者与败者。mark_conflict=true 记录冲突,apply=true 自动将非保护败方标记为 superseded。 |
| memory_list_conflictsC | 列出记忆冲突记录,默认只看 open 状态。 |
| memory_confirmB | 将一条记忆标记为用户确认,提升为 user_confirmed + locked 保护级别,禁止自动覆盖。 |
| memory_statusA | 查看 memory-arbiter 运行状态:数据库路径、降级模式、客户端标识、策略配置。 |
| memory_audit_summaryA | 返回各 workspace 的记忆统计概览:条目数、最旧/最新条目时间、open 冲突数、各 source_type 分布。纯 SQL 聚合,不做语义判断,用于快速判断是否需要深入审查。 |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
Latest Blog Posts
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/billy12151/memory-arbiter-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server