skills.rest
SullivanXiong
SullivanXiongCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

dedupe-entities

Identify duplicate Person entities in Memory MCP and merge confirmed clusters.

Updated Aug 23, 2026
Search Tags:#knowledge graph#mcp#memory management#deduplication#data cleaning#entity resolution
One-click install
npx skills add https://github.com/SullivanXiong/dotfiles --skill dedupe-entities
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: dedupe-entities
Source: https://github.com/SullivanXiong/dotfiles/tree/main/home/.claude/skills/dedupe-entities
Command: npx skills add https://github.com/SullivanXiong/dotfiles --skill dedupe-entities

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill addresses the issue of fragmented knowledge within the Memory MCP by identifying and merging duplicate 'Person' entities that represent the same individual, ensuring a cleaner and more accurate knowledge base.

Core Features & Use Cases

  • Duplicate Detection: Identifies duplicate 'Person' entities based on shared identifiers (GitHub username, email, Slack ID) or similar names.
  • Candidate Presentation: Clearly presents potential duplicate clusters, including the entities, the reason for matching, and a recommended canonical entity.
  • User-Confirmed Merging: Requires explicit user confirmation before executing any merges, ensuring data integrity.
  • Use Case: Regularly run this skill to clean up your personal knowledge graph, merging scattered information about contacts into single, comprehensive profiles.

Quick Start

Run the dedupe entities skill to clean up your memory graph.

Frequently Asked Questions about dedupe-entities

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I deduplicate person entities in a Memory MCP knowledge graph?▼

Entity resolution in a knowledge graph works by identifying 'Person' entities with shared identifiers like GitHub usernames or emails, clustering them, presenting candidates for user confirmation, and migrating observations and relations to a canonical entity upon approval.

How do I merge duplicate contacts in my Memory MCP knowledge base?▼

To clean your memory graph, run the deduplication skill to execute Python scripts for graph analysis, review the identified duplicate 'Person' clusters alongside their matching reasons, and confirm the merges to consolidate scattered contact information into comprehensive profiles.

Can I automate entity resolution without user confirmation in the Memory MCP?▼

No, deduplication requires explicit user confirmation before executing any merges, ensuring data integrity by allowing you to manually review duplicate clusters and the recommended canonical entity before observations and relations are migrated.

Do I need Python to perform data cleaning on Memory MCP entities?▼

Yes, you need Python scripts to perform data cleaning on Memory MCP entities because the deduplication process relies on these scripts for graph analysis, while utilizing MCP tools for the actual memory operations to migrate relations.

What are the limitations of using name similarity for entity resolution in a knowledge graph?▼

Using name similarity for entity resolution can generate false positives, which is why the system presents duplicate clusters with matching reasons and a recommended canonical entity for explicit user confirmation before merging any observations.

Related Skills

Explore More Data & Analytics Skills→
zhengxuyuzhengxuyu

ontology

Manage typed knowledge graphs with Python scripts and a JSONL store.

Community
Advanced
gaos6egaos6e

ontology

Create and validate typed knowledge graphs of entities and relations.

Community
Advanced
semantica-agisemantica-agi
3.0k

deduplicate

Detect and merge duplicate entities, groups, and relationships in Semantica knowledge graphs.

Official
Intermediate
ethereumclassicethereumclassic
12

videogen

Orchestrate call processing pipelines from transcripts to rendered video content.

Official
Advanced