Menu
Multi-layer memory system for persistent storage and semantic search.
Instalar com Codex ou Claude Copie este prompt, cole no Codex, Claude ou outro assistente e deixe que ele revise a página da skill e instale para você.
Baseado na classificação ocupacional SOC
| name | memory |
| description | Multi-layer memory system for persistent storage and semantic search. |
Persistent memory system with semantic search, PARA organization, and multi-user isolation.
Search across all memory sources using hybrid vector + keyword search.
Parameters:
query (string, required): Search querymaxResults (number, optional, default: 10): Maximum results to returnminScore (number, optional, default: 0.3): Minimum relevance scoreExample:
{
"query": "project deadline",
"maxResults": 5
}
Read a specific memory file or section.
Parameters:
relPath (string, required): Relative path to the memory filefrom (number, optional): Start line numberlines (number, optional): Number of lines to readExample:
{
"relPath": "MEMORY.md",
"from": 1,
"lines": 50
}
Log a fact to the daily log for future reference.
Parameters:
content (string, required): The fact to logExample:
{
"content": "User prefers morning meetings"
}
Add information to the auto-memory system.
Parameters:
category (string, required): Memory category
preferences - User preferenceslearnings - Learned informationdecisions - Decisions madecontext - Context about current situationnotes - General notescontent (string, required): Content to adduserId (string, optional): User-specific memoryExample:
{
"category": "preferences",
"content": "Timezone: America/Chicago"
}
Create a structured entity in PARA organization.
Parameters:
name (string, required): Entity namebucket (string, required): PARA bucket
projects - Active projectsareas - Ongoing areasresources - Reference resourcesarchives - Inactive itemsExample:
{
"name": "Q1 Roadmap",
"bucket": "projects"
}
Add an atomic fact to a PARA entity.
Parameters:
entityPath (string, required): Path to the entityfact (string, required): The fact to addcategory (string, required): Fact category
milestone - Key milestonesstatus - Current statusrelationship - Relationshipspreference - Preferencesuser-signal - User signalsExample:
{
"entityPath": "projects/Q1 Roadmap",
"fact": "Launch target: March 30",
"category": "milestone"
}
| Source | Description |
|---|---|
memory | Shared MEMORY.md and memory/*.md files |
user | Per-user isolated memory |
conversation | Conversation history |
logFact for chronological trackingaddMemory to store user preferences and contextHTTP endpoints, SSE streaming, MCP tool servers, and API design patterns for Hybrid. Use when designing new endpoints, streaming responses, or integrating with the agent protocol.
Deep knowledge of Hybrid's agent runtime architecture, including agent server, memory system, scheduler, and channel adapters. Use when working on core agent infrastructure, understanding data flow, or debugging system-level issues.
OpenClaw template format, memory conventions, and compatibility requirements for Hybrid. Use when working with agent templates, memory files, or ensuring OpenClaw feature parity.
Microservices patterns, container architecture, deployment targets (Fly.io, Cloudflare Workers), and system design decisions for Hybrid. Use when designing deployments, scaling strategies, or container configurations.
TypeScript strict mode patterns, branded types, discriminated unions, and code style conventions used in Hybrid. Use when writing or reviewing TypeScript code, defining types, or enforcing code quality.
UI/UX patterns, documentation site conventions, and design system for Hybrid's documentation and mini apps. Use when working on the site, mini apps, or user-facing interfaces.