The Memory Research skill enables AI agents to perform comprehensive web searches, synthesize technical and organizational data, and maintain a structured knowledge graph.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install memory-research
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install memory-research using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Memory Research Skill?
This skill empowers AI agents to go beyond their training data by performing real-time web research on companies, individuals, technologies, and specific topics. It acts as a bridge between raw internet data and a structured personal knowledge management (PKM) system, ensuring that every piece of information is verified, summarized, and stored as a reusable entity. By integrating this into Openclaw Skills, developers can ensure their agents maintain a living, breathing database of external subjects.
Memory Research is designed to handle both explicit triggers, like direct requests for information, and implicit triggers, such as a user providing a bare URL or a company name. The skill prioritizes accuracy by using hedging language for unverified public data and emphasizes the importance of user context, ensuring that the stored knowledge is not just factual but personally relevant to the user's workflow.
Memory Research Skill Use Cases
- Researching a new technology or framework to understand its ecosystem, maturity, and alternatives.
- Conducting due diligence on organizations, including funding stages, leadership teams, and core products.
- Profiling individuals to capture their professional background, current roles, and areas of expertise.
- Synthesizing complex domain topics into actionable summaries for a long-term knowledge graph.
- Automating the transition from a raw URL or name to a structured memory entity within Openclaw Skills.
How Memory Research Skill Works
- Initiates web research using multiple search queries to gather a well-rounded perspective from current, diverse sources.
- Queries existing Basic Memory to prevent duplication and determines whether to update an existing note or propose a new one.
- Generates a structured summary categorized by entity type (Organization, Person, Tech, or Topic) with appropriate hedging for uncertain data.
- Presents synthesized findings to the user for approval, allowing for modifications and the addition of personal context.
- Executes the creation of a Markdown-formatted note with specific metadata, directory placement, and taxonomy tags.
- Establishes relational links between the new entity and existing knowledge to build a dense, interconnected graph within Openclaw Skills.
Memory Research Skill Setup
To integrate this skill into your agent's workflow, ensure you have an active search tool and the Basic Memory module configured. Copy the skill definition into your agent's skill directory:
# Navigate to your agent's configuration directory
mkdir -p .ai/skills
# Add the memory-research skill definition
# Ensure your agent has permission to use search_notes and write_note tools
Memory Research Skill Data Schema & Taxonomy
The skill organizes data into a structured hierarchy within the Basic Memory system. Each entity is stored as a Markdown file with specific metadata:
| Entity Type | Directory | Note Type | Key Metadata Fields |
|---|---|---|---|
| Organization | organizations |
organization |
Stage, HQ, Employees, Leadership, Products |
| Person | people |
person |
Role, Expertise, Affiliations, Notable Work |
| Technology | concepts |
concept |
Maintainer, Maturity, License, Alternatives |
| Topic | concepts |
concept |
Definition, Trends, Key Players, Relevance |
Memory Research Skill Advanced Features
- Multi-source synthesis to avoid reliance on single sources and improve data reliability.
- User context capture to append unique observations, such as where the user first encountered the subject.
- Automated relationship mapping (e.g., 'works_at' or 'alternative_to') between entities in the knowledge graph.
- Dynamic template adaptation based on the detected entity type for consistent data formatting.
- Support for Openclaw Skills hooks to trigger updates when new information about a saved entity is discovered during separate tasks.
SKILL.md
Loading
Related Openclaw Skills
Miaomiao AI Assistant
A versatile multi-functional AI agent that handles everything from real-time news and weather updates to image generation and data visualization.
QQBot Media Sender
A specialized skill for automating the transmission of images, videos, and documents to QQ chat windows with built-in optimization.
Best Skill Recommendations
A sophisticated recommendation engine that evaluates, compares, and installs the most effective tools for your AI agent while ensuring system compatibility.
Gcal Manager
A local, credential-free reference tool providing instant access to Gcal Manager documentation, patterns, and implementation guides.
Solscan Pro Data Skill
A professional-grade Solana blockchain data connector for AI agents to query accounts, transactions, tokens, and DeFi analytics via the Solscan Pro API.
Metadata Naming
A rigorous naming convention standard designed to turn disorganized file systems into stable, parseable, and searchable metadata-driven architectures.
Featured*
