skills.rest
omeraltn
omeraltnCOMMUNITY
Explore all skills by this author →
GitHub

agency-identity-graph-operator

Resolve records to canonical entity identifiers with deterministic blocking and field-level scoring.

Updated Apr 11, 2026
Search Tags:#multi-agent#deduplication#audit-trail#matching#tenant-isolation#identity-resolution#entity-graph
One-click install
npx skills add https://github.com/omeraltn/ice_cream_website_testing --skill agency-identity-graph-operator-omeraltn
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agency-identity-graph-operator
Source: https://github.com/omeraltn/ice_cream_website_testing/tree/main/.antigravity/agency-identity-graph-operator
Command: npx skills add https://github.com/omeraltn/ice_cream_website_testing --skill agency-identity-graph-operator-omeraltn

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

When multiple agents ingest or act on the same real-world entities from different sources, they often create duplicate records, conflicting actions, and cascading errors. This Skill provides a deterministic, evidence-driven identity layer so every agent resolves to the same canonical entity_id and avoids duplicate billing, duplicate shipments, and inconsistent customer state.

Core Features & Use Cases

  • Deterministic Resolution: Normalize fields, block candidates, compute field-level scores, and return a canonical entity_id with a confidence score and per-field evidence.
  • Collaborative Proposals & Auditing: When confidence is ambiguous, generate merge or split proposals with explicit evidence and maintain a full audit trail of decisions and agent provenance.
  • Graph Integrity & Safety: Support optimistic locking, simulation/previews of mutations, tenant-scoped queries, PII masking, and rollback for erroneous merges.
  • Use Case: In an e-commerce platform, prevent double-charges and duplicate shipments by resolving orders and customers from multiple ingestion points (web, support, third-party integrations) into a single canonical entity.

Quick Start

Resolve this record against the shared identity graph and return the canonical entity_id, confidence, per-field evidence, and whether to auto-merge or propose.

Frequently Asked Questions about agency-identity-graph-operator

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

FAQPage Schema
How do I prevent duplicate customer records when multiple agents ingest data from different sources?▼

Identity resolution prevents duplicate records by normalizing fields, deterministically blocking candidates, and computing field-level scores to return a single canonical entity_id for every agent action. This ensures shared state across disparate ingestion points.

What's the best way to resolve conflicting entity actions in a multi-agent system?▼

Entity graph resolution resolves conflicting actions by mapping records to canonical identifiers with confidence scores and per-field evidence. It generates explicit merge or split proposals with full audit logging when agent confidence is ambiguous.

How does deterministic identity resolution handle ambiguous entity matches?▼

Deterministic identity resolution handles ambiguous matches by generating evidence-rich merge or split proposals rather than auto-committing. It maintains an audit trail of decisions and agent provenance to ensure graph integrity and rollback safety.

Can I use tenant-scoped queries and PII masking with an entity identity graph?▼

Entity identity graph operations support tenant-scoped queries and PII masking to ensure data isolation. Optimistic locking, simulation previews, and rollback for erroneous merges maintain graph integrity across multi-tenant environments.

Why do I need an identity layer for multi-agent e-commerce integrations?▼

An identity layer prevents double-charges and duplicate shipments by resolving orders and customers from web, support, and third-party integrations into a single canonical entity. This guarantees consistent customer state across all agents.

Does multi-agent identity resolution work without external dependencies?▼

Multi-agent identity resolution operates without external dependencies by performing deterministic blocking, normalization, and field-level scoring internally. It provides confidence scores and per-field evidence directly within the agent workflow.

Related Skills

Explore More Software Engineering Skills→
AtemndobsAtemndobs

RFP Source Expansion

Normalize federal, state, and local RFP streams into a unified data model.

Community
Advanced
rajyeole6rajyeole6

agency-identity-graph-operator

Resolve disparate records into canonical entities within a multi-agent system.

Community
Advanced
AreteDriverAreteDriver
4

entity-resolver

Resolves entity ambiguity across document corpora using alias-aware, confidence-scored matching and consolidation into canonical forms.

Community
Advanced
wosnatwosnat

cypher-queries

Run Cypher queries against a Neo4j knowledge graph using prepared templates.

Community
Advanced